lanthanide and actinide chemistry · ffx ffx jwbk057-fm jwbk057-cotton december 9, 2005 14:5 char...

FFX FFX

JWBK057-FM JWBK057-Cotton December 9, 2005 14:5 Char Count=

Lanthanide and ActinideChemistry

Simon CottonUppingham School, Uppingham, Rutland, UK

iii

FFX FFX


Lanthanide and Actinide Chemistry

i

FFX FFX


Inorganic Chemistry

A Wiley Series of Advanced Textbooks

Editorial Board

Derek Woollins, University of St. Andrews, UKBob Crabtree, Yale University, USADavid Atwood, University of Kentucky, USAGerd Meyer, University of Hannover, Germany

Previously Published Books In this Series

Chemical Bonds: A DialogAuthor: J. K. Burdett

Bioinorganic Chemistry: Inorganic Elements in the Chemistryof Life – An Introduction and GuideAuthor: W. Kaim

Synthesis of Organometallic Compounds: A Practical GuideEdited by: S. Komiya

Main Group Chemistry Second EditionAuthor: A. G. Massey

Inorganic Structural ChemistryAuthor: U. Muller

Stereochemistry of Coordination CompoundsAuthor: A. Von Zelewsky

Lanthanide and Actinide ChemistryAuthor: S. A. Cotton

ii

FFX FFX


Lanthanide and ActinideChemistry

Simon CottonUppingham School, Uppingham, Rutland, UK

iii

FFX FFX


Copyright C© 2006 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,West Sussex PO19 8SQ, EnglandTelephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected] our Home Page on www.wileyeurope.com or www.wiley.com

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means,electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without thepermission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd,The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (+44) 1243 770620.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in thisbook are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with anyproduct or vendor mentioned in this book.

This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on theunderstanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required,the services of a competent professional should be sought.

Other Wiley Editorial Offices

John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA

Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA

Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany

John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809

John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.

Library of Congress Cataloging-in-Publication Data

Cotton, Simon, Dr.Lanthanide and actinide chemistry / Simon Cotton.

p. cm. – (Inorganic chemistry)Includes bibliographical references and index.ISBN-13: 978-0-470-01005-1 (acid-free paper)ISBN-10: 0-470-01005-3 (acid-free paper)ISBN-13: 978-0-470-01006-8 (pbk. : acid-free paper)ISBN-10: 0-470-01006-1 (pbk. : acid-free paper)1. Rare earth metals. 2. Actinide elements. I. Title.

II. Inorganic chemistry (John Wiley & Sons)QD172.R2C68 2006546′.41—dc22 2005025297

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN-13 9-78-0-470-01005-1 (Cloth) 9-78-0-470-01006-8 (Paper)ISBN-10 0-470-01005-3 (Cloth) 0-470-01006-1 (Paper)

Typeset in 10/12pt Times by TechBooks, New Delhi, IndiaPrinted and bound in Great Britain by Antony Rowe, Chippenham, WiltshireThis book is printed on acid-free paper responsibly manufactured from sustainable forestryin which at least two trees are planted for each one used for paper production.

iv

http://www.wileyeurope.com

http://www.wiley.com

FFX FFX


In memory of Ray and Derek Cotton, my parents.

Remember that it was of your parents you were born;how can you repay what they have given to you?

(Ecclesiasticus 7.28 RSV)

also in memory of Marıa de los Angeles Santiago Hernandez,a lovely lady and devout Catholic, who died far too young.

and to Lisa.

v

FFX FFX


vi

FFX FFX


Dr Simon Cotton obtained his PhD at Imperial College London. After postdoctoral re-search and teaching appointments at Queen Mary College, London, and the University ofEast Anglia, he has taught chemistry in several different schools, and has been at Upping-ham School since 1996. From 1984 until 1997, he was Editor of Lanthanide and ActinideCompounds for the Dictionary of Organometallic Compounds and the Dictionary of Inor-ganic Compounds. He authored the account of Lanthanide Coordination Chemistry for the2nd edition of Comprehensive Coordination Chemistry (Pergamon) as well as the accountsof Lanthanide Inorganic and Coordination Chemistry for both the 1st and 2nd editions ofthe Encyclopedia of Inorganic Chemistry (Wiley).

His previous books are:

S.A. Cotton and F.A. Hart, “The Heavy Transition Elements”, Macmillan, 1975.D.J. Cardin, S.A. Cotton, M. Green and J.A. Labinger, “Organometallic Compounds of the

Lanthanides, Actinides and Early Transition Metals”, Chapman and Hall, 1985.S.A. Cotton, “Lanthanides and Actinides”, Macmillan, 1991.S.A. Cotton, “Chemistry of Precious Metals”, Blackie, 1997.

vii

FFX FFX


Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

1 Introduction to the Lanthanides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Characteristics of the Lanthanides . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 The Occurrence and Abundance of the Lanthanides . . . . . . . . . . . . . . . . . . 21.4 Lanthanide Ores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.5 Extracting and Separating the Lanthanides . . . . . . . . . . . . . . . . . . . . . . 3

1.5.1 Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.5.2 Separating the Lanthanides . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.6 The Position of the Lanthanides in the Periodic Table . . . . . . . . . . . . . . . . . 61.7 The Lanthanide Contraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 The Lanthanides – Principles and Energetics . . . . . . . . . . . . . . . . . . . . . . 92.1 Electron Configurations of the Lanthanides and f Orbitals . . . . . . . . . . . . . . . 92.2 What do f Orbitals Look Like? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.3 How f Orbitals affect Properties of the Lanthanides . . . . . . . . . . . . . . . . . . 102.4 The Lanthanide Contraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.5 Electron Configurations of the Lanthanide Elements and of Common Ions . . . . . . . 122.6 Patterns in Ionization Energies . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.7 Atomic and Ionic Radii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.8 Patterns in Hydration Energies (Enthalpies) for the Lanthanide Ions . . . . . . . . . . 142.9 Enthalpy Changes for the Formation of Simple Lanthanide Compounds . . . . . . . . 14

2.9.1 Stability of Tetrahalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.9.2 Stability of Dihalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.9.3 Stability of Aqua Ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.10 Patterns in Redox Potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3 The Lanthanide Elements and Simple Binary Compounds . . . . . . . . . . . . . . . . 233.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.2 The Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.2.1 Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.2.2 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.2.3 Alloys and Uses of the Metals . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.3 Binary Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.3.1 Trihalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.3.2 Tetrahalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273.3.3 Dihalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.3.4 Oxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

ix

FFX FFX


x Contents

3.4 Borides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.5 Carbides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.6 Nitrides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.7 Hydrides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.8 Sulfides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

4 Coordination Chemistry of the Lanthanides . . . . . . . . . . . . . . . . . . . . . . . 354.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.2 Stability of Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.3 Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.3.1 The Aqua Ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.3.2 Hydrated Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.3.3 Other O-Donors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384.3.4 Complexes of β-Diketonates . . . . . . . . . . . . . . . . . . . . . . . . . 394.3.5 Lewis Base Adducts of β-Diketonate Complexes . . . . . . . . . . . . . . . . 414.3.6 Nitrate Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414.3.7 Crown Ether Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . 414.3.8 Complexes of EDTA and Related Ligands . . . . . . . . . . . . . . . . . . . 424.3.9 Complexes of N-Donors . . . . . . . . . . . . . . . . . . . . . . . . . . . 434.3.10 Complexes of Porphyrins and Related Systems . . . . . . . . . . . . . . . . . 444.3.11 Halide Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464.3.12 Complexes of S-Donors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

4.4 Alkoxides, Alkylamides and Related Substances . . . . . . . . . . . . . . . . . . . . 474.4.1 Alkylamides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474.4.2 Alkoxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484.4.3 Thiolates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

4.5 Coordination Numbers in Lanthanide Complexes . . . . . . . . . . . . . . . . . . . 504.5.1 General Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504.5.2 Examples of the Coordination Numbers . . . . . . . . . . . . . . . . . . . . 514.5.3 The Lanthanide Contraction and Coordination Numbers . . . . . . . . . . . . 534.5.4 Formulae and Coordination Numbers . . . . . . . . . . . . . . . . . . . . . 54

4.6 The Coordination Chemistry of the +2 and +4 States . . . . . . . . . . . . . . . . . 544.6.1 The (+2) State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554.6.2 The (+4) State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

5 Electronic and Magnetic Properties of the Lanthanides . . . . . . . . . . . . . . . . . 615.1 Magnetic and Spectroscopic Properties of the Ln3+ Ions . . . . . . . . . . . . . . . . 615.2 Magnetic Properties of the Ln3+ Ions . . . . . . . . . . . . . . . . . . . . . . . . . 62

5.2.1 Adiabatic Demagnetization . . . . . . . . . . . . . . . . . . . . . . . . . . 655.3 Energy Level Diagrams for the Lanthanide Ions, and their

Electronic Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 655.3.1 Electronic Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665.3.2 Hypersensitive Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . 68

5.4 Luminescence Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695.4.1 Quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735.4.2 Antenna Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735.4.3 Applications of Luminescence to Sensory Probes . . . . . . . . . . . . . . . . 745.4.4 Fluorescence and TV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 755.4.5 Lighting Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

FFX FFX


Contents xi

5.4.6 Lasers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765.4.7 Euro Banknotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

5.5 NMR Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775.5.1 β-Diketonates as NMR Shift Reagents . . . . . . . . . . . . . . . . . . . . . 775.5.2 Magnetic Resonance Imaging (MRI) . . . . . . . . . . . . . . . . . . . . . . 785.5.3 What Makes a Good MRI Agent? . . . . . . . . . . . . . . . . . . . . . . . 795.5.4 Texaphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

5.6 Electron Paramagnetic Resonance Spectroscopy . . . . . . . . . . . . . . . . . . . . 825.7 Lanthanides as Probes in Biological Systems . . . . . . . . . . . . . . . . . . . . . 83

6 Organometallic Chemistry of the Lanthanides . . . . . . . . . . . . . . . . . . . . . . 896.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 896.2 The +3 Oxidation State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

6.2.1 Alkyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 896.2.2 Aryls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

6.3 Cyclopentadienyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 916.3.1 Compounds of the Unsubstituted Cyclopentadienyl Ligand

(C5H5 = Cp; C5Me5 = Cp*) . . . . . . . . . . . . . . . . . . . . . . . . . 916.3.2 Compounds [LnCp*3] (Cp* = Pentamethylcyclopentadienyl) . . . . . . . . . . 946.3.3 Bis(cyclopentadienyl) Alkyls and Aryls LnCp2R . . . . . . . . . . . . . . . . 956.3.4 Bis(pentamethylcyclopentadienyl) Alkyls . . . . . . . . . . . . . . . . . . . 966.3.5 Hydride Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

6.4 Cyclooctatetraene Dianion Complexes . . . . . . . . . . . . . . . . . . . . . . . . 986.5 The +2 State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

6.5.1 Alkyls and Aryls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 996.5.2 Cyclopentadienyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1006.5.3 Other Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

6.6 The +4 State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1016.7 Metal–Arene Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1016.8 Carbonyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

7 The Misfits: Scandium, Yttrium, and Promethium . . . . . . . . . . . . . . . . . . . . 1077.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1077.2 Scandium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

7.2.1 Binary Compounds of Scandium . . . . . . . . . . . . . . . . . . . . . . . 1087.3 Coordination Compounds of Scandium . . . . . . . . . . . . . . . . . . . . . . . . 108

7.3.1 The Aqua Ion and Hydrated Salts . . . . . . . . . . . . . . . . . . . . . . . 1087.3.2 Other Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1097.3.3 Alkoxides and Alkylamides . . . . . . . . . . . . . . . . . . . . . . . . . . 1117.3.4 Patterns in Coordination Number . . . . . . . . . . . . . . . . . . . . . . . 112

7.4 Organometallic Compounds of Scandium . . . . . . . . . . . . . . . . . . . . . . . 1147.5 Yttrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147.6 Promethium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

8 The Lanthanides and Scandium in Organic Chemistry . . . . . . . . . . . . . . . . . . 1218.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1218.2 Cerium(IV) Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

8.2.1 Oxidation of Aromatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1218.2.2 Oxidation of Alkenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1238.2.3 Oxidation of Alcohols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

FFX FFX


xii Contents

8.3 Samarium(II) Iodide, SmI2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1248.3.1 Reduction of Halogen Compounds . . . . . . . . . . . . . . . . . . . . . . . 1248.3.2 Reduction of α-Heterosubstituted Ketones . . . . . . . . . . . . . . . . . . . 1248.3.3 Reductions of Carbonyl Groups . . . . . . . . . . . . . . . . . . . . . . . . 1248.3.4 Barbier Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1268.3.5 Reformatsky Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

8.4 Lanthanide β-Diketonates as Diels–Alder Catalysts . . . . . . . . . . . . . . . . . . 1278.4.1 Hetero-Diels–Alder Reactions . . . . . . . . . . . . . . . . . . . . . . . . . 128

8.5 Cerium(III) Chloride and Organocerium Compounds . . . . . . . . . . . . . . . . . 1288.6 Cerium(III) Chloride and Metal Hydrides . . . . . . . . . . . . . . . . . . . . . . . 1308.7 Scandium Triflate and Lanthanide Triflates . . . . . . . . . . . . . . . . . . . . . . 131

8.7.1 Friedel–Crafts Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . 1318.7.2 Diels–Alder Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1328.7.3 Mannich Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1328.7.4 Imino-Diels–Alder Reactions . . . . . . . . . . . . . . . . . . . . . . . . . 133

8.8 Alkoxides and Aryloxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1348.9 Lanthanide Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1358.10 Organometallics and Catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

9 Introduction to the Actinides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1459.1 Introduction and Occurrence of the Actinides . . . . . . . . . . . . . . . . . . . . . 1459.2 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1459.3 Extraction of Th, Pa, and U . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

9.3.1 Extraction of Thorium . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1479.3.2 Extraction of Protactinium . . . . . . . . . . . . . . . . . . . . . . . . . . 1489.3.3 Extraction and Purification of Uranium . . . . . . . . . . . . . . . . . . . . 148

9.4 Uranium Isotope Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1489.4.1 Gaseous Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1489.4.2 Gas Centrifuge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1489.4.3 Electromagnetic Separation . . . . . . . . . . . . . . . . . . . . . . . . . . 1499.4.4 Laser Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

9.5 Characteristics of the Actinides . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1499.6 Reduction Potentials of the Actinides . . . . . . . . . . . . . . . . . . . . . . . . . 1509.7 Relativistic Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

10 Binary Compounds of the Actinides . . . . . . . . . . . . . . . . . . . . . . . . . . . 15510.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15510.2 Halides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

10.2.1 Syntheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15610.2.2 Structure Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

10.3 Thorium Halides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15910.4 Uranium Halides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

10.4.1 Uranium(VI) Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . 16110.4.2 Uranium(V) Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . 16110.4.3 Uranium (IV) Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . 16110.4.4 Uranium(III) Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . 16310.4.5 Uranium Hexafluoride and Isotope Separation . . . . . . . . . . . . . . . . . 163

FFX FFX


Contents xiii

10.5 The Actinide Halides (Ac–Am) excluding U and Th . . . . . . . . . . . . . . . . . 16510.5.1 Actinium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16510.5.2 Protactinium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16510.5.3 Neptunium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16610.5.4 Plutonium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16610.5.5 Americium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

10.6 Halides of the Heavier Transactinides . . . . . . . . . . . . . . . . . . . . . . . . 16810.6.1 Curium(III) Chloride . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16810.6.2 Californium(III) Chloride, Californium(III) Iodide, and Californium(II) Iodide . 16810.6.3 Einsteinium(III) Chloride . . . . . . . . . . . . . . . . . . . . . . . . . . 168

10.7 Oxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16910.7.1 Thorium Oxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16910.7.2 Uranium Oxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

10.8 Uranium Hydride UH3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17010.9 Oxyhalides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

11 Coordination Chemistry of the Actinides . . . . . . . . . . . . . . . . . . . . . . . . . 17311.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17311.2 General Patterns in the Coordination Chemistry of the Actinides . . . . . . . . . . . 17311.3 Coordination Numbers in Actinide Complexes . . . . . . . . . . . . . . . . . . . . 17411.4 Types of Complex Formed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17411.5 Uranium and Thorium Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . 175

11.5.1 Uranyl Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17511.5.2 Coordination Numbers and Geometries in Uranyl Complexes . . . . . . . . 17711.5.3 Some Other Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . 17811.5.4 Uranyl Nitrate and its Complexes; their Role in Processing Nuclear Waste . . 17911.5.5 Nuclear Waste Processing . . . . . . . . . . . . . . . . . . . . . . . . . 179

11.6 Complexes of the Actinide(IV) Nitrates and Halides . . . . . . . . . . . . . . . . . 18011.6.1 Thorium Nitrate Complexes . . . . . . . . . . . . . . . . . . . . . . . . 18011.6.2 Uranium(IV) Nitrate Complexes . . . . . . . . . . . . . . . . . . . . . . 18011.6.3 Complexes of the Actinide(IV) Halides . . . . . . . . . . . . . . . . . . . 181

11.7 Thiocyanates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18311.8 Amides, Alkoxides and Thiolates . . . . . . . . . . . . . . . . . . . . . . . . . . 184

11.8.1 Amide Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18411.8.2 Alkoxides and Aryloxides . . . . . . . . . . . . . . . . . . . . . . . . . 185

11.9 Chemistry of Actinium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18611.10 Chemistry of Protactinium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18711.11 Chemistry of Neptunium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

11.11.1 Complexes of Neptunium . . . . . . . . . . . . . . . . . . . . . . . . . 18811.12 Chemistry of Plutonium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

11.12.1 Aqueous Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18911.12.2 The Stability of the Oxidation States of Plutonium . . . . . . . . . . . . . . 19011.12.3 Coordination Chemistry of Plutonium . . . . . . . . . . . . . . . . . . . 19111.12.4 Plutonium in the Environment . . . . . . . . . . . . . . . . . . . . . . . 193

11.13 Chemistry of Americium and Subsequent Actinides . . . . . . . . . . . . . . . . . 19511.13.1 Potentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

11.14 Chemistry of the Later Actinides . . . . . . . . . . . . . . . . . . . . . . . . . . 196

FFX FFX


xiv Contents

12 Electronic and Magnetic Properties of the Actinides . . . . . . . . . . . . . . . . . . . 20112.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20112.2 Absorption Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

12.2.1 Uranium(VI) – UO22+ – f 0 . . . . . . . . . . . . . . . . . . . . . . . . . 202

12.2.2 Uranium(V) – f 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20312.2.3 Uranium(IV) – f 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20412.2.4 Spectra of the Later Actinides . . . . . . . . . . . . . . . . . . . . . . . 206

12.3 Magnetic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

13 Organometallic Chemistry of the Actinides . . . . . . . . . . . . . . . . . . . . . . . 20913.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20913.2 Simple σ-Bonded Organometallics . . . . . . . . . . . . . . . . . . . . . . . . . 20913.3 Cyclopentadienyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

13.3.1 Oxidation State (VI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21113.3.2 Oxidation State (V) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21113.3.3 Oxidation State (IV) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21113.3.4 Oxidation State (III) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

13.4 Compounds of the Pentamethylcyclopentadienyl Ligand (C5Me5 = Cp*) . . . . . . . 21513.4.1 Oxidation State(IV) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21513.4.2 Cationic Species and Catalysts . . . . . . . . . . . . . . . . . . . . . . . 21613.4.3 Hydrides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21713.4.4 Oxidation State (III) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218

13.5 Tris(pentamethylcyclopentadienyl) Systems . . . . . . . . . . . . . . . . . . . . . 21913.6 Other Metallacycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21913.7 Cyclooctatetraene Dianion Compounds . . . . . . . . . . . . . . . . . . . . . . . 21913.8 Arene Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22113.9 Carbonyls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

14 Synthesis of the Transactinides and their Chemistry . . . . . . . . . . . . . . . . . . . 22514.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22514.2 Finding New Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22614.3 Synthesis of the Transactinides . . . . . . . . . . . . . . . . . . . . . . . . . . . 22614.4 Naming the Transactinides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22914.5 Predicting Electronic Arrangements . . . . . . . . . . . . . . . . . . . . . . . . . 23014.6 Identifying the Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23014.7 Predicting Chemistry of the Transactinides . . . . . . . . . . . . . . . . . . . . . 23314.8 What is known about the Chemistry of the Transactinides . . . . . . . . . . . . . . 234

14.8.1 Element 104 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23414.8.2 Element 105 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23414.8.3 Element 106 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23414.8.4 Element 107 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23514.8.5 Element 108 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

14.9 And the Future? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253

FFX FFX


Preface

This book is aimed at providing a sound introduction to the chemistry of the lanthanides,actinides and transactinides to undergraduate students. I hope that it will also be of valueto teachers of these courses. Whilst not being anything resembling a comprehensive mono-graph, it does attempt to give a factual basis to the area, and the reader can use a fairlycomprehensive bibliography to range further.

Since I wrote a previous book in this area (1991), the reader may wonder why on earth Ihave bothered again. The world of f-block chemistry has moved on. It is one of active andimportant research, with names like Bunzli, Evans, Ephritikhine, Lappert, Marks and Parkerfamiliar world-wide (I am conscious of names omitted). Not only have several more elementsbeen synthesized (and claims made for others), but lanthanides and their compounds areroutinely employed in many areas of synthetic organic chemistry; gadolinium compoundsfind routine application in MRI scans; and there are other spectroscopic applications, notablyin luminescence. Whilst some areas are hardly changed, at this level at least (e.g. actinidemagnetism and spectroscopy), a lot more compounds have been described, accounting forthe length of the chapters on coordination and organometallic chemistry. I have tried tospell out the energetics of lanthanide chemistry in more detail, whilst I have provided someend-of-chapter questions, of variable difficulty, which may prove useful for tutorials. I havesupplied most, but not all, of the answers to these (my answers, which are not alwaysdefinitive).

It is a pleasure to thank all those who have contributed to the book: Professor DerekWoollins, for much encouragement at different stages of the project; Professor JamesAnderson, for many valuable comments on Chapter 8; Martyn Berry, who supplied valuablecomment on early versions of several chapters; to Professors Michel Ephritikhine, AllanWhite and Jack Harrowfield, and Dr J.A.G. Williams, and many others, for exchanginge-mails, correspondence and ideas. I’m very grateful to Dr Mary P. Neu for much informa-tion on plutonium. The staff of the Libraries of the Chemistry Department of CambridgeUniversity and of the Royal Society of Chemistry, as well as the British Library, havebeen quite indispensable in helping with access to the primary literature. I would also wishto thank a number of friends – once again Dr Alan Hart, who got me interested in lan-thanides in the first place; Professor James Anderson (again), Dr Andrew Platt, Dr JohnFawcett, and Professor Paul Raithby, for continued research collaboration and obtainingspectra and structures from unpromising crystals, so that I have kept a toe-hold in the area.Over the last 8 years, a number of Uppingham 6th form students have contributed to myefforts in lanthanide coordination chemistry – John Bower, Oliver Noy, Rachel How, ViliusFranckevicius, Leon Catallo, Franz Niecknig, Victoria Fisher, Alex Tait and Joanna Harris.Finally, thanks are most certainly due to Dom Paul-Emmanuel Clenet and the Benedictinecommunity of the Abbey of Bec, for continued hospitality during several Augusts when Ihave been compiling the book.

Simon Cotton

xv

FFX FFX


xvi

SPH SPH

JWBK057-01 JWBK057-Cotton December 7, 2005 19:54 Char Count=

1 Introduction to the Lanthanides

By the end of this chapter you should be able to:

� understand that lanthanides differ in their properties from the s- and d-block metals;� recall characteristic properties of these elements;� appreciate reasons for their positioning in the Periodic Table;� understand how the size of the lanthanide ions affects certain properties and how this can

be used in the extraction and separation of the elements;� understand how to obtain pure samples of individual Ln3+ ions.

1.1 Introduction

Lanthanide chemistry started in Scandinavia. In 1794 Johann Gadolin succeeded in ob-taining an ‘earth’(oxide) from a black mineral subsequently known as gadolinite; he calledthe earth yttria. Soon afterwards, M.H. Klaproth, J.J. Berzelius and W. Hisinger obtainedceria, another earth, from cerite. However, it was not until 1839–1843 that the Swede C.G.Mosander first separated these earths into their component oxides; thus ceria was resolvedinto the oxides of cerium and lanthanum and a mixed oxide ‘didymia’ (a mixture of theoxides of the metals from Pr through Gd). The original yttria was similarly separated intosubstances called erbia, terbia, and yttria (though some 40 years later, the first two nameswere to be reversed!). This kind of confusion was made worse by the fact that the newlydiscovered means of spectroscopic analysis permitted misidentifications, so that around 70‘new’ elements were erroneously claimed in the course of the century.

Nor was Mendeleev’s revolutionary Periodic Table a help. When he first published hisPeriodic Table in 1869, he was able to include only lanthanum, cerium, didymium (nowknown to have been a mixture of Pr and Nd), another mixture in the form of erbia, andyttrium; unreliable information about atomic mass made correct positioning of these ele-ments in the table difficult. Some had not yet been isolated as elements. There was no wayof predicting how many of these elements there would be until Henry Moseley (1887–1915)analysed the X-ray spectra of elements and gave meaning to the concept of atomic number.He showed that there were 15 elements from lanthanum to lutetium (which had only beenidentified in 1907). The discovery of radioactive promethium had to wait until after WorldWar 2.

It was the pronounced similarity of the lanthanides to each other, especially each to itsneighbours (a consequence of their general adoption of the +3 oxidation state in aqueoussolution), that caused their classification and eventual separation to be an extremely difficultundertaking.

Lanthanide and Actinide Chemistry S. CottonC© 2006 John Wiley & Sons, Ltd.

SPH SPH



Subsequently it was not until the work of Bohr and of Moseley that it was known preciselyhow many of these elements there were. Most current versions of the Periodic Table placelanthanum under scandium and yttrium.

1.2 Characteristics of the Lanthanides

The lanthanides exhibit a number of features in their chemistry that differentiate them fromthe d-block metals. The reactivity of the elements is greater than that of the transition metals,akin to the Group II metals:

1. A very wide range of coordination numbers (generally 6–12, but numbers of 2, 3 or 4are known).

2. Coordination geometries are determined by ligand steric factors rather than crystal fieldeffects.

3. They form labile ‘ionic’ complexes that undergo facile exchange of ligand.4. The 4f orbitals in the Ln3+ ion do not participate directly in bonding, being well shielded

by the 5s2 and 5p6 orbitals. Their spectroscopic and magnetic properties are thus largelyuninfluenced by the ligand.

5. Small crystal-field splittings and very sharp electronic spectra in comparison with thed-block metals.

6. They prefer anionic ligands with donor atoms of rather high electronegativity (e.g.O, F).

7. They readily form hydrated complexes (on account of the high hydration energy of thesmall Ln3+ ion) and this can cause uncertainty in assigning coordination numbers.

8. Insoluble hydroxides precipitate at neutral pH unless complexing agents are present.9. The chemistry is largely that of one (3+) oxidation state (certainly in aqueous solution).

10. They do not form Ln=O or Ln≡N multiple bonds of the type known for many transitionmetals and certain actinides.

11. Unlike the transition metals, they do not form stable carbonyls and have (virtually) nochemistry in the 0 oxidation state.

1.3 The Occurrence and Abundance of the Lanthanides

Table 1.1 presents the abundance of the lanthanides in the earth’s crust and in the solarsystem as a whole. (Although not in the same units, the values in each list are internallyconsistent.)

Two patterns emerge from these data. First, that the lighter lanthanides are more abundantthan the heavier ones; secondly, that the elements with even atomic number are moreabundant than those with odd atomic number. Overall, cerium, the most abundant lanthanide

Table 1.1 Abundance of the lanthanides

La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y

Crust (ppm) 35 66 9.1 40 0.0 7 2.1 6.1 1.2 4.5 1.3 3.5 0.5 3.1 0.8 31Solar System 4.5 1.2 1.7 8.5 0.0 2.5 1.0 3.3 0.6 3.9 0.9 2.5 0.4 2.4 0.4 40.0(with respect to

107 atoms Si)

SPH SPH


Extracting and Separating the Lanthanides 3

on earth, has a similar crustal concentration to the lighter Ni and Cu, whilst even Tm andLu, the rarest lanthanides, are more abundant than Bi, Ag or the platinum metals.

The abundances are a consequence of how the elements were synthesized by atomic fusionin the cores of stars with heavy elements only made in supernovae. Synthesis of heaviernuclei requires higher temperature and pressures and so gets progressively harder as theatomic number increases. The odd/even alternation (often referred to as the Oddo–Harkinsrule) is again general, and reflects the facts that elements with odd mass numbers have largernuclear capture cross sections and are more likely to take up another neutron, so elementswith odd atomic number (and hence odd mass number) are less common than those witheven mass number. Even-atomic-number nuclei are more stable when formed.

1.4 Lanthanide Ores

Principal sources (Table 1.2) are the following:Bastnasite LnFCO3; Monazite (Ln, Th)PO4 (richer in earlier lanthanides); Xenotime(Y, Ln)PO4 (richer in later lanthanides). In addition to these, there are Chinese rare earthreserves which amount to over 70% of the known world total, mainly in the form of theionic ores from southern provinces. These Chinese ion-absorption ores, weathered graniteswith lanthanides adsorbed onto the surface of aluminium silicates, are in some cases low incerium and rich in the heavier lanthanides (Longnan) whilst the Xunwu deposits are rich inthe lighter metals; the small particle size makes them easy to mine. The Chinese ores havemade them a leading player in lanthanide chemistry.

Table 1.2 Typical abundance of the lanthanides in oresa

% La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y

Monazite 20 43 4.5 16 0 3 0.1 1.5 0.05 0.6 0.05 0.2 0.02 0.1 0.02 2.5Bastnasite 33.2 49.1 4.3 12 0 0.8 0.12 0.17 160 310 50 35 8 6 1 0.1Xenotime 0.5 5 0.7 2.2 0 1.9 0.2 4 1 8.6 2 5.4 0.9 6.2 0.4 60.0

a Bold values are in ppm.

1.5 Extracting and Separating the Lanthanides

These two processes are not necessarily coterminous. Whilst electronic, optical and mag-netic applications require individual pure lanthanides, the greatest quantity of lanthanidesis used as mixtures, e.g. in mischmetal or oxide catalysts.

1.5.1 Extraction

After initial concentration by crushing, grinding and froth flotation, bastnasite is treated with10% HCl to remove calcite, by which time the mixture contains around 70% lanthanideoxides. This is roasted to oxidize the cerium content to CeIV; on further extraction withHCl, the Ce remains as CeO2, whilst the lanthanides in the (+3) state dissolve as a solutionof the chlorides.

Monazite is usually treated with NaOH at 150 ◦C to remove phosphate as Na3PO4, leavinga mixture of the hydrated oxides, which are dissolved in boiling HCl at pH 3.5, separating thelanthanides from insoluble ThO2. Sulfuric acid can also be used to dissolve the lanthanides.

SPH SPH



1.5.2 Separating the Lanthanides

These can be divided into four types: chemical separations, fractional crystallization, ion-exchange methods and solvent extraction. Of these, only the last-named is used on a com-mercial scale (apart from initial separation of cerium). Chemical separations rely on usingstabilities of unusual oxidation states; thus Eu2+ is the only ion in that oxidation state formedon reduction by zinc amalgam and can then be precipitated as EuSO4 (note the similaritywith heavier Group 2 metals). Repeated (and tedious) fractional crystallization, which madeuse of slight solubility differences between the salts of neighbouring lanthanides, such asthe bromates Ln(BrO3)3.9H2O, ethyl sulfates and double nitrates, were once the only pos-sible way of obtaining pure lanthanides, as with the 15 000 recrystallizations carried outby the American C. James to get pure thulium bromate (1911) (Figure 1.1 indicates theprinciple of this method).

A1

A2B1

C1

D1

E1 D2 C3 B4 A5

B5C4D3E2

C2 B3 A4

B2 A3

Mos

t sol

uble

cons

titue

nt

Least soluble

constituent

originalsolutionLegend

Middle fraction

=====

dissolve in H2Oevaporate to crystalsmother liquorcrystalscombine

Figure 1.1Diagrammatic representation of the system of fractional crystallization used to separate salts of therare-earth elements (reproduced with permission from D.M. Yost, H. Russell and C.S. Garner, TheRare Earth Elements and their Compounds, John Wiley, 1947.)

Ion-exchange chromatography is not of real commercial importance for large-scale pro-duction, but historically it was the method by which fast high-purity separation of thelanthanides first became feasible. As radioactive lanthanide isotopes are important fissionproducts of the fission of 235U and therefore need to be separated from uranium, and be-cause the actinides after plutonium tend to resemble the lanthanides, the development ofthe technique followed on the Manhattan project. It was found that if Ln3+ ions were ad-sorbed at the top of a cation-exchange resin, then treated with a complexing agent such asbuffered citric acid, then the cations tended to be eluted in reverse atomic number order(Figure 1.2a); the anionic ligand binds most strongly to the heaviest (and smallest) cation,which has the highest charge density. A disadvantage of this approach when scaled up tohigh concentration is that the peaks tend to overlap (Figure 1.2b).

It was subsequently found that amine polycarboxylates such as EDTA4− gave strongercomplexes and much better separations. In practice, some Cu2+ ions (‘retainer’) areadded to prevent precipitation of either the free acid H4EDTA or the lanthanide complex

SPH SPH


Extracting and Separating the Lanthanides 5

Sm

Nd

Pr

150

00 60

Con

cent

ratio

n –

mg.

oxi

de/l

iter

Con

cent

ratio

n –

cou

nts

/min

.

101

105

Y

Tb

DyTm

HoLu

25000Elution Time – min. Volume of Eluate – liters

(a) (b)

Figure 1.2(a) Cation-exchange chromatography of lanthanides, (b) overlap of peaks at high concentration. (a) Tracer-scale elutionwith 5% citrate at pH 3.20 (redrawn from B.H. Ketelle and G.E. Boyd, J. Am. Chem. Soc., 1947, 69, 2800). (b) Macro-scaleelution with 0.1% citrate at pH 5.30 (redrawn from F.H. Spedding, E.I. Fulmer, J.E. Powell, and T.A. Butler, J. Am. Chem.Soc., 1950, 72, 2354). Reprinted with permission of the American Chemical Society c©1978.

HLn(EDTA).xH2O on the resin. The major disadvantage of this method is that it is a slowprocess for large-scale separations.

Solvent extraction has come to be used for the initial stage of the separation process, togive material with up to 99.9% purity. In 1949, it was found that Ce4+ could readily beseparated from Ln3+ ions by extraction from a solution in nitric acid into tributyl phosphate[(BuO)3PO]. Subsequently the process was extended to separating the lanthanides, using anon-polar organic solvent such as kerosene and an extractant such as (BuO)3PO or bis(2-ethylhexyl)phosphinic acid [[C4H9CH(C2H5)]2P=O(OH)] to extract the lanthanidesfrom aqueous nitrate solutions. The heavier lanthanides form complexes which are moresoluble in the aqueous layer. After the two immiscible solvents have been agitated togetherand separated, the organic layer is treated with acid and the lanthanide extracted. The solventis recycled and the aqueous layer put through further stages.

For a lanthanide LnA distributed between two phases, a distribution coefficent DA isdefined:

DA = [LnA in organic phase] / [LnA in aqueous phase]

For two lanthanides LnA and LnB in a mixture being separated, a separation factor βAB can

be defined, where

βAB = DA/DB

β is very close to unity for two adjacent lanthanides in the Periodic Table (obviously, thelarger β is, the better the separation).

In practice this process is run using an automated continuous counter-current circuit inwhich the organic solvent flows in the opposite direction to the aqueous layer containingthe lanthanides. An equilibrium is set up between the lanthanide ions in the aqueous phaseand the organic layer, with there tending to be a relative enhancement of the concentrationof the heavier lanthanides in the organic layer. Because the separation between adjacent

SPH SPH



Back-extract

Back extract Back-extract

Back extraction10 stages

Back extraction

Back extraction

Aqueoussolution Sm-Nd

Aqueoussolution Pr-La

AqueousNd

AqueousPr

AqueousLa

AqueousSm, etc.

(water)

Stripped

Strippedsolvent Stripped

solvent

solventfor

re-cycling

SolventcontainingSm-Nd-HNO3

SolventSm

SolventPr

Solvent

Precipitateand

re-dissolve

Concentrateor

precipitateand

re-dissolve

Scrub Scrub

Scrub

Scrub and Extraction

Scrub and extractionPerhaps 20 stages

Scrub and extraction

Feed SolventFeed

SolventFeed5M HNO3 Saturated

neutral nitratesolution ofLa-Pr-Nd-Sm

(Undilutedtributyl phosphate)

Figure 1.3Schematic diagram of lanthanon separation by solvent extraction. From R.J. Callow, The Rare EarthIndustry, Pergamon, 1966; reproduced by permission.

lanthanides in each exchange is relatively slight, over a thousand exchanges are used (seeFigure 1.3). This method affords lanthanides of purity up to the 99.9% purity level and isthus well suited to large-scale separation, the products being suited to ordinary chemicaluse. However, for electronic or spectroscopic use (‘phosphor grade’) 99.999% purity isnecessary, and currently ion-exchange is used for final purification to these levels. Thedesired lanthanides are precipitated as the oxalate or hydroxide and converted into theoxides (the standard starting material for many syntheses) by thermal decomposition.

Various other separation methods have been described, one recent one involving the useof supercritical carbon dioxide at 40 ◦C and 100 atm to convert the lanthanides into theircarbonates whilst the quadrivalent metals (e.g. Th and Ce) remain as their oxides.

1.6 The Position of the Lanthanides in the Periodic Table

As already mentioned, neither Mendeleev nor his successors could ‘place’ the lanthanides inthe Periodic Table. Not only was there no recognizable atomic theory until many years after-wards, but, more relevant to how groupings of elements were made in those days, there wasno comparable block of elements for making comparisons. The lanthanides were sui generis.The problem was solved by the combined (but separate) efforts of Moseley and Bohr, the for-mer showing that La–Lu was composed of 15 elements with atomic numbers from 57 to 71,whilst the latter concluded that the fourth quantum shell could accommodate 32 electrons,and that the lanthanides were associated with placing electrons into the 4f orbitals.

SPH SPH


The Lanthanide Contraction 7

The Periodic Table places elements in atomic number order, with the lanthanides fallingbetween barium (56) and hafnium (72). For reasons of space, most present-day PeriodicTables are presented with Groups IIA and IVB (2 and 4) separated only by the Group IIIB(3) elements. Normally La (and Ac) are grouped with Sc and Y, but arguments have beenadvanced for an alternative format, in which Lu (and Lr) are grouped with Sc and Y (seee.g. W.B. Jensen, J. Chem. Educ., 1982, 59, 634) on the grounds that trends in properties(e.g. atomic radius, I.E., melting point) in the block Sc-Y-Lu parallel those in the GroupTi-Zr-Hf rather closely, and that there are resemblances in the structures of certain binarycompounds. Certainly on size grounds, Lu resembles Y and Sc (it is intermediate in sizebetween them) rather more than does La, owing to the effects of the ‘lanthanide contraction’.The resemblances between Sc and Lu are, however, by no means complete.

1.7 The Lanthanide Contraction

The basic concept is that there is a decrease in radius of the lanthanide ion Ln3+ on crossingthe series from La to Lu. This is caused by the poor screening of the 4f electrons. This causesneighbouring lanthanides to have similar, but not identical, properties, and is discussed inmore detail in Section 2.4.

Question 1.1 Using the information you have been given in Section 1.2, draw up a tablecomparing (in three columns) the characteristic features of the s-block metals (use group 1as typical) and the d-block transition metals.Answer 1.1 see Table 1.3 for one such comparison.

Table 1.3 Comparison of 4f, 3d and Group I metals

4f 3d Group I

Electron configurationsof ions

Variable Variable Noble gas

Stable oxidation states Usually +3 Variable 1Coordination numbers in

complexesCommonly 8–10 Usually 6 Often 4–6

Coordination polyhedra incomplexes

Minimise repulsion Directional Minimise repulsion

Trends in coordinationnumbers

Often constant in block Often constant in block Increase down group

Donor atoms in complexes ‘Hard’ preferred ‘Hard’ and ‘soft’ ‘Hard’ preferredHydration energy High Usually moderate LowLigand exchange reactions Usually fast Fast and slow FastMagnetic properties of

ionsIndependent of

environmentDepends on environment

and ligand fieldNone

Electronic spectra of ions Sharp lines Broad lines NoneCrystal field effects in

complexesWeak Strong None

Organometalliccompounds

Usually ionic, some withcovalent character

Covalently bonded Ionically bonded

Organometallics in lowoxidation states

Few Common None

Multiply bonded atoms incomplexes

None Common None

SPH SPH


8

SPH SPH


2 The Lanthanides – Principlesand Energetics


� recognise the difference between f-orbitals and other types of orbitals;� understand that they are responsible for the particular properties of the lanthanides;� give the electron configurations of the lanthanide elements and Ln3+ ions;� explain the reason for the lanthanide contraction;� understand the effect of the lanthanide contraction upon properties of the lanthanides and

subsequent elements;� explain patterns in properties such as ionization and hydration energies;� recall that lanthanides behave similarly when there is no change in the 4f electron pop-

ulation, but that they differ when the change involves a change in the number of 4felectrons;

� relate the stability of oxidation states to the ionization energies;� calculate enthalpy changes for the formation of the aqua ions and of the lanthanide halides

and relate these to the stability of particular compounds.

2.1 Electron Configurations of the Lanthanides and f Orbitals

The lanthanides (and actinides) are those in which the 4f (and 5f) orbitals are gradually filled.At lanthanum, the 5d subshell is lower in energy than 4f, so lanthanum has the electronconfiguration [Xe] 6s2 5d1 (Table 2.1).

As more protons are added to the nucleus, the 4f orbitals contract rapidly and becomemore stable than the 5d (as the 4f orbitals penetrate the ‘xenon core’ more) (see Figure 2.1),so that Ce has the electron configuration [Xe] 6s2 5d1 4f1 and the trend continues with Prhaving the arrangement [Xe] 6 s2 4f 3. This pattern continues for the metals Nd–Eu, all ofwhich have configurations [Xe] 6s2 4f n (n = 4–7) After europium, the stability of the half-filled f subshell is such that the next electron is added to the 5d orbital, Gd being [Xe] 6s2

5d14f 7; at terbium, however, the earlier pattern is resumed, with Tb having the configuration[Xe] 6s2 4f 9, and succeeding elements to ytterbium being [Xe] 6s2 4f n (n = 10–14).The last lanthanide, lutetium, where the 4f subshell is now filled, is predictably [Xe] 6s2

5d14f 14.


SPH SPH


10 The Lanthanides – Principles and Energetics

Table 2.1 Electron configurations of the lanthanides and their common ions

Atom Ln3+ Ln4+ Ln2+

La [Xe] 5d1 6s2 [Xe]Ce [Xe] 4f1 5d1 6s2 [Xe] 4f1 [Xe]Pr [Xe] 4f3 6s2 [Xe] 4f2 [Xe] 4f1

Nd [Xe] 4f4 6s2 [Xe] 4f3 [Xe] 4f2 [Xe] 4f4

Pm [Xe] 4f5 6s2 [Xe] 4f4

Sm [Xe] 4f6 6s2 [Xe] 4f5 [Xe] 4f6

Eu [Xe] 4f7 6s2 [Xe] 4f6 [Xe] 4f7

Gd [Xe] 4f7 5d1 6s2 [Xe] 4f7

Tb [Xe] 4f9 6s2 [Xe] 4f8 [Xe] 4f7

Dy [Xe] 4f10 6s2 [Xe] 4f9 [Xe] 4f8 [Xe] 4f10

Ho [Xe] 4f11 6s2 [Xe] 4f10

Er [Xe] 4f12 6s2 [Xe] 4f11

Tm [Xe] 4f13 6s2 [Xe] 4f12 [Xe] 4f13

Yb [Xe] 4f14 6s2 [Xe] 4f13 [Xe] 4f14

Lu [Xe] 4f14 5d1 6s2 [Xe] 4f14

Y [Kr] 4d1 5s2 [Kr]

Pro

babi

lity

4f

5d

6s

r

Figure 2.1The radial part of the hydrogenic wave functions for the 4f, 5d and 6s orbitals of cerium (after H.G.Friedman et al. J. Chem. Educ. 1964, 41, 357). Reproduced by permission of the American ChemicalSociety c© 1964.

2.2 What do f Orbitals Look Like?

They are generally represented in one of two ways, either as a cubic set, or as a generalset, depending upon which way the orbitals are combined. The cubic set comprises fxyz;fz(x2−y2), fz(y2−z2) and fy(z2−x2); fz3, fx3 and fy3.

The general set, more useful in non-cubic environments, uses a different combination:fz3; fxz2 and fyz2; fxyz; fz(x2−y2), fx(x2−3y2) and fy(3x2−y2); Figure 2.2 shows the general set.

2.3 How f Orbitals affect Properties of the Lanthanides

The 4f orbitals penetrate the xenon core appreciably. Because of this, they cannot overlapwith ligand orbitals and therefore do not participate significantly in bonding. As a result of

SPH SPH


The Lanthanide Contraction 11

z

y

x

++

+

+

x

(a) fz 3

(c) fx(x2−3y 2)

+

+

(b) fxz2

(d) fxyz

z

x

−−

−

−

−

− −

+

− +

y

x

z

Figure 2.2(a) fz3 , (fx3 and fy3 are similar, extending along the x- and y-axes repectively); (b) fxz3 , (fyz3 is similar,produced by a 90◦ rotation about the z-axis); (c) fx(x2−3y2), fy(3x2−y2) is similar, formed by a 90◦

clockwise rotation round the z-axis); (d) fxyz , (fxz2−y2 ), fy(z2−y2) and fz(x2−y2) are produced by a 45◦

rotation about the x , y and z-axes respectively). The cubic set comprises fx3 , fy3 , fz3 , fxyz , fx(z2−y2)fy(z2−x2) and fz(x2−y2); the general set is made of fz3 , fxz2 , fyz2 , fxyz , fz(x2−y2), fx(x2−3y2) and fy(3x2−y2).(Reproduces with permission from S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991).

their isolation from the influence of the ligands, crystal-field effects are very small (and canbe regarded as a perturbation on the free-ion states) and thus electronic spectra and magneticproperties are essentially unaffected by environment. The ability to form π bonds is alsoabsent, and thus there are none of the M=O or M≡N bonds found for transition metals (or,indeed, certain early actinides). The organometallic chemistry is appreciably different fromthat of transition metals, too.

2.4 The Lanthanide Contraction

As the series La–Lu is traversed, there is a decrease in both the atomic radii and in the radii ofthe Ln3+ ions, more markedly at the start of the series. The 4f electrons are ‘inside’ the 5s and5p electrons and are core-like in their behaviour, being shielded from the ligands, thus takingno part in bonding, and having spectroscopic and magnetic properties largely independentof environment. The 5s and 5p orbitals penetrate the 4f subshell and are not shielded from

SPH SPH



increasing nuclear charge, and hence because of the increasing effective nuclear chargethey contract as the atomic number increases. Some part (but only a small fraction) of thiseffect has also been ascribed to relativistic effects. The lanthanide contraction is sometimesspoken of as if it were unique. It is not, at least in the way that the term is usually used.Not only does a similar phenomenon take place with the actinides (and here relativisticeffects are much more responsible) but contractions are similarly noticed on crossing thefirst and second long periods (Li–Ne; Na–Ar) not to mention the d-block transition series.However, as will be seen, because of a combination of circumstances, the lanthanides adoptprimarily the (+3) oxidation state in their compounds, and therefore demonstrate the steadyand subtle changes in properties in a way that is not observed in other blocks of elements.

The lanthanide contraction has a knock-on effect in the elements in the 5d transitionseries. It would naturally be expected that the 5d elements would show a similar increasein size over the 4d transition elements to that which the 4d elements demonstrate over the3d metals. However, it transpires that the ‘lanthanide contraction’ cancels this out, almostexactly, and this has pronounced effects on the chemistry, e.g. Pd resembling Pt rather thanNi, Hf is extremely similar to Zr.

2.5 Electron Configurations of the Lanthanide Elements and of Common Ions

The electronic arrangements of the lanthanide atoms have already been mentioned (Sec-tion 2.1 and Table 2.1) where it can be seen that in general the ECs of the atoms are [Xe]4f n 5d0 6s2, the exceptions being La and Ce, where the 4f orbitals have not contractedsufficiently to bring the energy of the 4f electrons below that of the 5d electrons; Gd, wherethe effect of the half-filled 4f subshell dominates; and Lu, the 4f subshell having alreadybeen filled at Yb.

In forming the ions, electrons are removed first from the 6s and 5d orbitals (ratherreminiscent of the case of the transition metals, where they are removed from 4s beforethey are taken from 3d), so that all the Ln3+ ions have [Xe] 4f n arrangements.

2.6 Patterns in Ionization Energies

The values of the first four ionization energies for the lanthanides (and yttrium) are listedin Table 2.2.

As usual, for a particular element, I4 > I3 > I2 > I1, as the electron being removed isbeing taken from an ion with an increasingly positive charge, affording greater electrostaticattraction. Yttrium has greater ionization energies than the lanthanides as there is one fewerfilled shell, and decreased distance effects outweigh the effect of the reduced nuclear charge.

There is in general a tendency for ionization energies to increase on crossing the seriesbut it is irregular. The low I3 values for gadolinium and lutetium, where the one electronremoved comes from a d orbital, not an f orbital, and the high I3 value for Eu and Yb showsome correlation with the stabilizing effects of half-filled and filled f sub shells. Resultscan be presented diagrammatically as cumulative ionization energies (Figure 2.3). Thesediagrams demonstrate at a glance the effect of the alternately high and low values of I3 forthe elements Eu and Gd, influencing the respective stabilities of the +2 and +3 states ofthese elements, similarly illustrated by the neighbours Yb and Lu. The low I4 value for Ce(and to some extent for Pr and Tb) can be correlated with the accessibility of the Ce4+ ion.

SPH SPH


Patterns in Ionization Energies 13

Table 2.2 Ionization Energies (kJ/mole)

I1 I2 I3 I4 I1 + I2 I1 + I2 + I3 I1 + I2 + I3 + I4

La 538 1067 1850 4819 1605 3455 8274Ce 527 1047 1949 3547 1574 3523 7070Pr 523 1018 2086 3761 1541 3627 7388Nd 529 1035 2130 3899 1564 3694 7593Pm 536 1052 2150 3970 1588 3738 7708Sm 543 1068 2260 3990 1611 3871 7990Eu 546 1085 2404 4110 1631 4035 8145Gd 593 1167 1990 4250 1760 3750 8000Tb 564 1112 2114 3839 1676 3790 7629Dy 572 1126 2200 4001 1698 3898 7899Ho 581 1139 2204 4110 1720 3924 8034Er 589 1151 2194 4115 1740 3934 8049Tm 597 1163 2285 4119 1760 4045 8164Yb 603 1176 2415 4220 1779 4194 8414Lu 523 1340 2033 4360 1863 3896 8256Y 616 1181 1980 5963 1797 3777 9740

Figure 2.3Cumulative ionization energies across the lanthanide Series (reproduced by permission of Macmillanfrom S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991).

SPH SPH



2.7 Atomic and Ionic Radii

These are listed in Table 2.3 and shown in Figure 2.4. It will be seen that the atomicradii exhibit a smooth trend across the series with the exception of the elements europiumand ytterbium. Otherwise the lanthanides have atomic radii intermediate between thoseof barium in Group 2A and hafnium in Group 4A, as expected if they are representedas Ln3+ (e−)3. Because the screening ability of the f electrons is poor, the effective nu-clear charge experienced by the outer electrons increases with increasing atomic number,so that the atomic radius would be expected to decrease, as is observed. Eu and Yb areexceptions to this; because of the tendency of these elements to adopt the (+2) state, theyhave the structure [Ln2+(e−)2] with consequently greater radii, rather similar to barium.In contrast, the ionic radii of the Ln3+ ions exhibit a smooth decrease as the series iscrossed.

The patterns in radii exemplify a principle enunciated by D.A. Johnson: ‘The lanthanideelements behave similarly in reactions in which the 4f electrons are conserved, and verydifferently in reactions in which the number of 4f electrons change’ (J. Chem. Educ., 1980,57, 475).

Table 2.3 Atomic and ionic radii of the lanthanides (pm)

Ba La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf

217.3 187.7 182.5 182.8 182.1 181.0 180.2 204.2 180.2 178.2 177.3 176.6 175.7 174.6 194.0 173.4 156.4La3+ Ce3+ Pr3+ Nd3+ Pm3+ Sm3+ Eu3+ Gd3+ Tb3+ Dy3+ Ho3+ Er3+ Tm3+ Yb3+ Lu3+ Y3+

103.2 101.0 99.0 98.3 97.0 95.8 94.7 93.8 92.3 91.2 90.1 89.0 88.0 86.8 86.1 90.0

2.8 Patterns in Hydration Energies (Enthalpies) for the Lanthanide Ions

Table 2.4 shows the hydration energies (enthalpies) for all the 3+ lanthanide ions, and alsovalues for the stablest ions in other oxidation states. Hydration energies fall into a patternLn4+ > Ln3+ > Ln2+, which can simply be explained on the basis of electrostatic attraction,

210

190

170

150

130

110

90La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Ln3+ radius/pmmetalic radius/pm

Rad

ius

(pm

)

Figure 2.4Metallic and ionic radii across the lanthanide series.

SPH SPH


Enthalpy Changes for the Formation of Simple Lanthanide Compounds 15

since the ions with a larger charge have a greater charge density. The hydration energiesfor the Ln3+ ions show a pattern of smooth increase with increasing atomic number, as theions becomes smaller and their attraction for water molecules increases. This is anotherexample of the principle enunciated by Johnson.

2.9 Enthalpy Changes for the Formation of Simple Lanthanide Compounds

Observed patterns of chemical behaviour can sometimes appear confusing or sometimesjust be encapsulated in rules [e.g. Ce has a stable (+4) oxidation state]. They can frequentlybe explained in terms of the energy changes involved in the processes.

2.9.1 Stability of Tetrahalides

Among the fluorides, lanthanum forms only LaF3, whilst cerium forms CeF3 and CeF4.(Tetrafluorides are also known for Pr and Tb, see Section 3.4). Why do neighbouring metalsbehave so differently? First, examine the energetics of formation of LaF3 using a Born–Haber cycle.

�H (kJ/mol)

La(s) → La(g) +402

La(g) → La3+(g) +538 + 1067 + 1850

3/2F2(g) → 3F (g) +252

3F(g) + 3 e → 3F−(g) −984

La3+(g) + 3F−(g) → LaF3 −4857

Thus La(s) + 3/2F2(g) → LaF3

�H = +402 + (538 + 1067 + 1850) + 252 − 984 − 4857 = −1732 kJ/mol

Calculated and observed enthalpies of formation for LnXn (n = 2–4) are given in Table 2.5.The same method can be used to calculate �Hf for LaF4, making the assumption that

the lattice energy is the same as that for CeF4 (−8391 kJ/mol), and that I4 for lanthanum is+4819 kJ/mol. In this case, �Hf(LaF4) = −691 kJ/mol. Similarly, using the same method,�Hf for LaF2 can be calculated as �H = −831 kJ/mol (assuming that the lattice energyis the same as for BaF2 (−2350 kJ/mol). This poses the question: if �Hf for LaF2 and forLaF4 are −831 and −691 kJ/mol respectively, why can’t these compounds be isolated?

Apart from the oversimplification of using �H rather than �G values as an index ofstability, what the preceding calculations have done is to indicate that these compoundsare stable with respect to the elements, and no other decomposition pathways have beenconsidered.

Table 2.4 Enthalpies of hydration of the lanthanide ions (values given as −�H hydr/kj mol−1

La3+ Ce3+ Pr3+ Nd3+ Pm3+ Sm3+ Eu3+ Gd3+ Tb3+ Dy3+ Ho3+ Er3+ Tm3+ Yb3+ Lu3+ Y3+

3278 3326 3373 3403 3427 3449 3501 3517 3559 3567 3623 3637 3664 3706 3722 3583Ce4+ Sm2+ Eu2+ Yb2+

6309 1444 1458 1594

SPH SPH



Table 2.5 Enthalpies of formation of lanthanide halidesa

LnF2 LnF3 LnF4 LnCI2 LnCI3 LnCI4 LnBr2 LnBr3 LnI2 LnI3

La 880 1732 600 520 1073 −480 430 907 320 699Ce 950 1733 1946 580 1058 820 490 890 380 686Pr 1050 1712 1690 700 1059 630 610 891 490 678Nd 1050 1661 1500 707 1042 490 630 873 510 665Pm 1080 1700 1500 720 1040 430 620 850 510 640Sm 1160 1669 1450 820 1040 400 720 857 600 640Eu 1188 1584 1290 824 1062 230 760 799 630 540Gd 870 1699 1310 500 937 180 400 829 290 619Tb 980 1707 1742 600 1008 600 500 850 390 598Dy 1050 1678 1540 693 1007 420 590 831 470 603Ho 1040 1698 1500 660 990 350 580 830 450 594Er 1020 1699 1510 640 995 360 560 836 420 586Tm 1090 1689 1500 709 995 360 640 840 500 582Yb 1172 1570 1250 799 960 250 710 800 580 561Lu 790 1640 1210 420 986 160 330 850 190 556

a Values are quoted as −�Hf (kJ/mol).Experimental values in bold.Values taken from D.W. Smith, J. Chem. Educ., 1986, 63, 228.

LaF4 might decompose thus :

LaF4 → LaF3 + 1/2 F2

Applying Hess’s Law to this in the form

�Hreaction = �g�H f (products) − ��H f (reactants)

�Hreaction = −1732 − (−691 + 0) = −1041 kJ/mol

This decomposition is thus thermodynamically favourable (especially as it would befavoured on entropy grounds too, with the formation of fluorine gas)

LaF2 might decompose by disproportionation:

3LaF2 → 2LaF3 + La

Using �Hf for LaF2 and for LaF3 (−831 and −1732 kJ/mol respectively), �H for thisdecomposition reaction can be calculated as −971 kJ/mol. This is again a very exothermicprocess, indicating that LaF2 is likely to be unstable (this is analogous to the reason forthe non-existence of MgCl, as disproportionation into Mg and MgCl2 is favoured, as thereader may be aware). The reason for this is that, although I3 for lanthanum is large (andendothermic), it is more than compensated for by the higher lattice energy for LaF3 comparedwith the value for LaF2.

Since both CeF3 and CeF4 are isolable, what makes the difference here? If similar calcu-lations are carried out (assuming lattice energies for CeF3 and CeF4 of −4915 and −8391kJ/mol respectively, and an enthalpy of atomization of 398 kJ/mol for Ce (see Table 2.6),and using the same enthalpy of atomization and electron affinity for fluorine as in the lan-thanum examples), �Hf for CeF3can be calculated as −1726 kJ/mol and �Hf for CeF4 as−1899 kJ/mol.

The discrepancy between �Hf for CeF3 and CeF4 is much smaller than is the case forlanthanum. In fact, for the decomposition reaction

CeF4 → geF3 + 1/2 F2

SPH SPH


Enthalpy Changes for the Formation of Simple Lanthanide Compounds 17

Table 2.6 Enthalpies of atomization of the lanthanides (kJ/mol)

Ba La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf

150.9 402.1 398 357 328 164.8 176 301 391 293 303 280 247 159 428.0 570.7

�H = −1726 − (−1899) = +173 kJ/mol, making the decomposition of CeF4 relativelyunfavourable. CeF4 is stable, certainly at ambient temperatures.

If the values of the parameters used in the calculation are compared, the determiningfactor is the much lower value of I4 for cerium (3547 kJ/mol compared with 4819 kJ/molfor La) due to the fact that the fourth electron is being removed from a different shell (nearerthe nucleus) in the case of lanthanum.

Other tetrahalides do not exist. Thus, though both CeCl3 and salts of the [CeCl6]2− ioncan be isolated, CeCl4 cannot be made. The reasons for this are those that enable fluorine tosupport high oxidation states (see Question 2.4). Similar factors indicate that tetrabromidesand tetraiodides are much less likely to be isolated.

2.9.2 Stability of Dihalides

The stability of the dihalides can be explained in a similar way. As already noted, LaF2 doesnot exist. When might dihalides be expected?

One way in which dihalides can decompose is disproportionation

3 LnX2 → 2 LnX3 + Ln

though a number of dihalides (particularly dichlorides) have been made by the reversereproportionation, by heating the mixture to a high temperature and rapidly quenching(Section 3.5.1.)

2 LnX3 + Ln → 3 LnX2

It can be shown, using Hess’s Law, that the disproportionation will be exothermic unless:

�Hf LnX3/�Hf LnX2 < 1.5

The disproportionation can be broken down into individual components (Figure 2.5)

�H1 = �H2 + �H3 + �H4

�H1 = [−3�Hlatt(LnX2)] + {2 I3(Ln) − [I1(Ln) + I2(Ln)]}+ [−�Hat(Ln) + 2 �Hlatt(LnX3)]

�H1 = {2 I3(Ln) − [I1(Ln) + I2(Ln)]} + 2 �Hlatt(LnX3) − 3 �Hlatt(LnX2) − �Hat(Ln)

∆H1

∆H3

∆H2 ∆H4

3 Ln2+(g) + 6X−(g) 2 Ln3+(g) + 6X−(g) + Ln(g)

2 LnX3(s) + Ln(s)3 LnX2(s)

Figure 2.5Disproportion of lanthanide dihalides.

SPH SPH



These equations can be used qualitatively, first to suggest for which lanthanides the halidesLnX2 are most likely to be stable. For the disproportionation process to be more likely to beendothermic (i.e. stabilizing the +2 state), the preceding equation suggests that high valuesof I3 are favourable. Study of Table 2.2 shows that this is more likely to be associated withEu, Sm and Yb. Iodide is the halide most often found in low oxidation state halides. In thecase of LnI2 , the large size of the iodide ion will reduce the lattice energy for both LnI2 andLnI3 so that the difference in lattice enthalpy will become less significant, favouring LnI2.The dihalide will also be favoured by a low �Hat(Ln), again associated with the lanthanidesmost often found in the +2 state, Eu and Yb (see Table 2.6).

Applying the �Hf LnX3/�Hf LnX2 <1.5 criterion, and using data in Table 2.5, thehalides LnX2 are most likely to be stable are predicted to be LnF2 (Ln = Sm, Eu, Yb);LnCl2(Ln = Nd, Pm, Sm, Eu, Dy, Tm, Yb); LnY2 (Y = Br, I; Ln = Pr, Nd, Pm, Sm, Eu,Dy, Ho, Er, Tm, Yb).

The known dihalides are listed in Table 3.2. There is a reasonably good correlation, giventhat the Pm dihalides have not been investigated on account of promethium’s short half-life.Some dihalides listed are ‘metallic’ and are not covered by this argument.

2.9.3 Stability of Aqua Ions

Since Ln3+(aq) is the most stable aqua ion, then both of the following processes are favoured.

Ln2+(aq) + H+(aq) → Ln3+(aq) + 1/2 H2(g)

and 2 Ln4+(aq) + H2O(aq) → 2 Ln3+(aq) + 2 H+(aq) + 1/2 O2(g)

In other words, Ln2+ ions tend to reduce water and Ln4+ ions tend to oxidize it. We canexamine the stability of the Ln2+ ion using a treatment similar to the one just employed(Figure 2.6), with

�Hox(Ln2+) = I3 + {[�Hhydr[Ln3+(aq)] − �Hhydr[Ln2+(aq)]

} − 439 kJ/mol.

When is it most likely that Ln2+(aq) ions will be stable? For the first of the two re-actions above to be favoured, the single factor that will help make �H positive is ahigh value of I3. Less important would be the size of the ions, as this could affect thehydration enthalpies; the difference between the hydration enthalpies will be less, thelarger the lanthanide ions. Substituting into the above equation, we can investigate therelative stabilities of La2+(aq) and Eu2+(aq), making use of ionization energies from

Ln2+(g)

Ln3+(g)

Ln2+(aq) + H+(aq)

Ln3+(aq) + 1/2H2(g)

∆Hhydr (Ln2+)

∆Hhydr (Ln3+)

I3 ∆HOX(Ln2+) + ∆HH

Figure 2.6Oxidation of Ln2+ (aq).

SPH SPH


Patterns in Redox Potentials 19

Table 2.2 and enthalpies of hydration found in Table 2.4, also assuming �Hhydr [(La2+(aq)] = −1327 kJ/mol:

For La2+:

�Hox(La2+) = I3 + {�Hhydr[La3+(aq)] − �Hhydr[La2+(aq)]} − 439

= 1850 + [−3278 − (−1327)] − 439

= −540 kJ/mol

For Eu2+:

�Hox(Eu2+) = I3 + {�Hhydr[Eu3+(aq)] − �Hhydr[Eu2+(aq)]} − 439

= 2404 + [−3501 − (−1458)] − 439

= −78 kJ/mol

The large exothermic value for �Hox (La2+) indicates that it is not likely to exist in aqueoussolution. The Eu2+(aq) ion is known to have a short lifetime in water, even though �His negative for the oxidation process, so the activation energy for oxidation may be ratherhigh.

2.10 Patterns in Redox Potentials

Known and estimated values are listed in Table 2.7. The values for the reduction potentialfor Ln3+ + 3 e− → Ln are very consistent, with slight irregularities at Eu and Yb. Thepotential largely depends upon three processes:

Ln(s) → Ln(g) �Hat(Ln)

Ln(g) → Ln3+(g) + 3 e− I1 + I2 + I3

Ln3+(g) → Ln3+(aq) �Hhydr(Ln3+)

The first two of these are endothermic and the third exothermic; overall �H is the differencebetween two large quantities. It remains fairly constant across the series, apart from Eu andYb, with values of 608 (La); 712 (Eu); 630 (Gd); 613 (Tm); 644 (Yb) and 593 (Lu) kJ/molbeing representative, the values for Eu and Yb resulting from the high I3 values. The verynegative value for the reduction potential is expected for such reactive metals (and alsoreflects the difficulty in isolating them). The potentials for the Ln3+ + e− → Ln2+ processreflect the stability of the +2 state. Since I3 relates to �H for the process, and �G = −nFE,a relationship between these is unsurprising. Similarly, the only potential for the Ln4++e− → Ln3+ process within reasonable range is that for Ce, and indicates that Ce4+ is theonly ion in this state likely to be encountered in aqueous solution.

Question 2.1 What is the trend in atomic radii, and that in ionic radii, for the lanthanides?What are the exceptions to this, and why?Answer 2.1 The structure of metals is usually described as one in which metal ions aresurrounded by a ‘sea’ of delocalised outer-shell electrons. The greater the number of looselyheld electrons, the stronger the metallic bonding and the smaller the atomic radius. If the

SPH SPH


Tabl

e2.

7R

edox

pote

ntia

lsof

the

lant

hani

deio

ns(V

)a

La

Ce

PrN

dPm

SmE

uG

dT

bD

yH

oE

rT

mY

bL

uY

Ln3+

+3e

→L

n−2

.37

−2.3

4−2

.35

−2.3

2−2

.29

−2.3

0−1

.99

−2.2

9−2

.30

−2.2

9−2

.33

−2.3

1−2

.31

−2.2

2−2

.30

−2.3

7L

n3++

e→

Ln2+

(−3.

1)(−

3.2)

(−2.

7)−2

.6b

(−2.

6)−1

.55

−0.3

4(−

3.9)

(−3.

7)−2

.5b

(−2.

9)(−

3.1)

−2.3

b−1

.05

Ln4+

+e

→L

n3+1.

70(3

.4)

(4.6

)(4

.9)

(5.2

)(6

.4)

(7.9

)(3

.3)

(5.0

)(6

.2)

(6.1

)(6

.1)

(7.1

)(8

.5)

aV

alue

sin

pare

nthe

ses

are

estim

ated

.b=

inT

HF.

20

SPH SPH


Patterns in Redox Potentials 21

lanthanides are represented as Ln3+ (e−)3, then the atomic radii would be expected to fallbetween those of barium [Ba2+(e−)2] and hafnium [Hf 4+(e−)4], as is generally observed.

Question 2.2 Using Hess’s Law and the �Hf for LaF2 (−831 kJ/mol) and for LaF3(−1732kJ/mol), calculate �H for this decomposition reaction.

3 LaF2 → 2 LaF3 + La

Answer 2.2

�Hreaction = ��Hf(products) − ��Hf(reactants)

�Hreaction = 2 (−1732) − 3 (−831 + 0) = −971 kJ/mol

Question 2.3 Assuming lattice energies for CeF3 and CeF4 of −4915 and −8391 kJ/molrespectively, and using the same enthalpy of atomization and electron affinity for fluorineas in the lanthanum examples (Section 2.9.1.), calculate �Hf for CeF3 and CeF4. Take anenthalpy of atomization of 398 kJ/mol for cerium (Table 2.6). Use ionization energies fromTable 2.2.Answer 2.3

�H (kJ/mol)

Ce(s) → Ce (g) +398

Ce(g) → Ce3+(g) +527 + 1047 + 1949

3/2 F2(g) → 3F(g) +252

3F(g) + 3e → 3F−(g) −984

Ce3+(g) + 3F−(g) → CeF3(s) −4915

Thus Ce(s) + 3/2 F2(g) → CeF3(s)

�H = +398 + (527 + 1047 + 1949) + 252 − 984 − 4915 = −1726 kJ/mol

�H (kJ/mol)

Ce(s) → Ce(g) +398

Ce(g) → Ce4+(g) +527 + 1047 + 1949 + 3547

2F2(g) → 4F (g) +336

4F(g) + 4e → 4F−(g) −1312

Ce4+(g) + 4F−(g) → CeF4(s) −8391

Thus Ce(s) + 2F2(g) → CeF4(s)

�H = +398 + (527 + 1047 + 1949 + 3547) + 336 − 1312 − 8391 = −1899 kJ/mol

Question 2.4 Unlike CeF4, CeCl4 does not exist, though CeCl3 does. Suggest why thismight be. Values of �Hf for CeCl3 and CeCl4 are −1058 and −820 (calculated) kJ/mol,respectively.Answer 2.4 Fluorine is well known to promote high oxidation states. Factors associatedwith this are the high lattice energies associated with the small fluoride ion, along with thevery small F–F bond energy (due to non-bonding electron pair repulsions) as well as highbond energies involving fluorine (not relevant in this case). Because of the larger size of theCl− ion, there is going to be much less difference between the lattice energies of CeCl3 and

SPH SPH



Key:

4

3

2

1

1800

2000

2200

2400

La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

= E 0

I 3(k

J m

ol −1

)

= I3

E0 (V

)

Figure 2.7Strong correlation between I3 and E◦ for Ln3+ + e → Ln2+ (reproduced by permission of Macmillanfrom S.A. Cotton, Lanthanides and Actinides, 1991, p. 26.)

CeCl4, and this is probably the determining factor. The higher �Hat(Cl) of 121.5 kJ/molalso mitigates against the formation of CeCl4.

Question 2.5 Applying the �HfLnX3/�HfLnX2 < 1.5 criterion for stability of dihalides,and using data in Table 2.5, predict which the halides LnX2 are most likely to be exist.Answer 2.5 As in section 2.9.2 LnF2(Ln = Sm, Eu, Yb); LnCl2(Ln = Nd, Pm, Sm, Eu,Dy, Tm, Yb); LnY2(Y = Br, I; Ln = Pr, Nd, Pm, Sm, Eu, Dy, Ho, Er, Tm, Yb) are predictedto be stable, in reasonably good agreement with the currently known facts.

Exercise 2.6 Using the same horizontal axis (atomic numbers 57–71), plot values of (a) thethird ionization enthalpy I3 and (b) the Ln3++ e → Ln2+ reduction potential on the y axis(choose appropriate scales). Comment.Answer 2.6 There is a strong correlation between them, not surprisingly! See Figure 2.7.

SPH SPH


3 The Lanthanide Elements andSimple Binary Compounds


� know how to prepare lanthanide metals and simple binary compounds such as the halides,oxides, and hydrides;

� apply the principle of the lanthanide contraction to explain patterns in co-ordinationnumber in these compounds;

� apply knowledge gained in the study of Chapter 2 to the compounds in unusual oxidationstates;

� understand the uses of the metals and certain compounds in applications such as hydrogenstorage and in superconductors.

3.1 Introduction

This chapter discusses the synthesis of the lanthanide metals, their properties, reactions, anduses. It also examines some of the most important binary compounds of the lanthanides,particularly the halides, which well illustrate patterns and trends in the lanthanide series.

3.2 The Elements

3.2.1 Properties

The lanthanides are rather soft reactive silvery solids with a metallic appearance, which tendto tarnish on exposure to air. They react slowly with cold water and rapidly in dilute acid.They ignite in oxygen at around 150–200 ◦C; similar reactions occur with the halogens,whilst they react on heating with many nonmetals such as hydrogen, sulfur, carbon, andnitrogen (above 1000 ◦C).

The metals are relatively high-melting and -boiling. Their physical properties usuallyshow smooth transition across the series, except that discontinuities are often observed forthe metals that have a stable +2 state, europium and ytterbium. Thus the atomic radii ofeuropium and ytterbium are about 0.2 A greater than might be predicted by interpolationfrom values for the flanking lanthanides (Figure 3.1).

Similarly, Sm, Eu, and Yb have boiling points that are lower than those of the neighbouringmetals (Figure 3.2).


SPH SPH


24 The Lanthanide Elements and Simple Binary Compounds

Figure 3.1The atomic radii of the lanthanide metals (reproduced with permission from S.A. Cotton, Lanthanidesand Actinides, Macmillan, 1991).

1000

2000

3000

4000


Bp

(K)

Figure 3.2The boiling points of the lanthanide metals (reproduced with permission from S.A. Cotton,Lanthanides and Actinides, Macmillan, 1991).

Assuming that one can represent the structure of a metal as a lattice of metal ions per-meated by a sea of electrons, then metals like lanthanum can be shown as (Ln3+) (e−)3;however, metals based upon divalent ions (like Eu and Yb) would be (Ln2+) (e−)2. The ionswith the (+3) charge have a smaller radius, as the higher charge draws in the electrons moreclosely, and the stronger attraction means that it takes more energy to boil them (similarly,they would be predicted to have higher conductivities).

3.2.2 Synthesis

The metals have similar reactivities to magnesium. This means that they cannot be extractedby methods like carbon reduction of the oxides; in practice, metallothermic reduction ofthe lanthanide fluoride or chloride with calcium at around 1450 ◦C is used. The product

SPH SPH


Binary Compounds 25

is an alloy of (excess of) calcium and the lanthanide, from which the calcium can bedistilled.

2 LnF3 + 3 Ca → 2 Ln + 3 CaF2

The reduction is carried out under an atmosphere of argon, not nitrogen.The above method is not suitable for obtaining the metals with a stable +2 state, which

are only reduced as far as the difluoride (Eu, Yb, Sm). The lanthanide can be removed bydistillation.

2 La + M2O3 � 2 M + La2O3

The ‘divalent’ metals Sm, Eu, and Yb have boiling points of 1791, 1597 and 1193 ◦Crespectively, much lower than that of La (3457 ◦C), so that on heating they are distilled off,their volatility meaning that their removal from the mixture will displace the equilibriumto the right, so the reaction will proceed to completion.

3.2.3 Alloys and Uses of the Metals

Mischmetal is the lanthanide alloy with the longest history. It is a mixture of the lighterlanthanides, cerium in particular, which is manufactured from an ‘unseparated’ mixtureof the oxides. This is first converted into a mixture of the anhydrous chlorides, which isthen electrolysed using a graphite anode and iron cathode at around 820 ◦C to afford themixture of metals. This is used mainly as an alloy with iron for the desulfurization anddeoxidation of steels and more familiarly in cigarette lighter flints. SmCo5 and other alloysof these metals have been used to make extremely strong permanent magnets. LaNi5 hasbeen widely examined as a material for hydrogen storage, with applications in fuel cells,catalytic hydrogenation, and removal of hydrogen from gas mixtures, as it rapidly absorbshydrogen at room temperature to afford compositions up to LaNi5H6; the hydrogen is givenup quickly at 140 ◦C. The most important application lies in rechargeable batteries for PCs(see Section 3.10).

3.3 Binary Compounds

3.3.1 Trihalides

Most halides are LnX3, but a number of LnX2 are known, as are a handful of tetrafluorides.

Syntheses of the Trihalides

Although the halides can be obtained as hydrates from reaction of the metal oxides orcarbonates with aqueous acids, these hydrates are hydrolysed on heating to the oxyhalideand thus the anhydrous halides (which are themselves deliquescent) cannot be made thatway.

LnX3 + H2O → LnOX + HX

The fluorides are obtained as insoluble hydrates LnF3.0.5H2O by precipitation and thehydrates dehydrated by heating in a current of anhydrous HF gas (or in vacuo).

LnF3.0.5H2O → LnF3 + 0.5 H2O

SPH SPH



Otherwise the anhydrous halides can generally be made by heating the metal with thehalogen (except for EuI3) or gaseous HCl.

2 Ln + 3 X2 = 2 LnX3

Another method for the chlorides involves refluxing the hydrated chlorides with thionyldichloride (SOCl2) for a few hours; an advantage of this method is that the other reactionproducts are gaseous SO2 and HCl.

LnCl3.xH2O + x SOCl2 → LnCl3 + x SO2 + 2x HCl

A route that works well in practice involves thermal decomposition of ammoniumhalogenometallates. Adding ammonium chloride to a solution of the metal oxides in hy-drochloric acid, followed by evaporation gives halogenometallate salts. These can be de-hydrated by heating with excess of ammonium chloride in a stream of gaseous HCl, theresulting anhydrous salt being decomposed by heating in vacuo at about 300 ◦C.

Ln2O3 + 9 NH4Cl + 3 HCl → 2 (NH4)3LnCl6 + 3 NH3 + 3 H2O

2 (NH4)3LnCl6 → 2 LnCl3 + 6 NH4Cl

The anhydrous halides can be purified by sublimation in vacuo, but owing to a tendency toreact with silica, contact with hot glass, thereby forming the oxyhalide, should be avoided.

Structures of the Trihalides

The lanthanide trihalides demonstrate very clearly the effect of varying the cation and anionradii upon the structure type adopted (Table 3.1).

The early lanthanide fluorides adopt the ‘LaF3’ structure (Figure 3.3) based on a metalion surrounded by a trigonal prism of fluorides with two additional capping fluorides, giving11 coordination (9 + 2), whilst the later fluorides have the ‘YF3’ structure. This is basedon tricapped trigonal prismatic 9 coordination, in which the prism is somewhat distorted.The structure adopted by UCl3 and several of the lanthanide halides is again a tricapped

Table 3.1 Structures of the lanthanide trihalides

F Cl Br I

La LaF3 (11) UCl3 (9) UCl3 (9) PuBr3 (8)Ce LaF3 (11) UCl3 (9) UCl3 (9) PuBr3 (8)Pr LaF3 (11) UCl3 (9) UCl3 (9) PuBr3 (8)Nd LaF3 (11) UCl3 (9) PuBr3 (8) PuBr3 (8)Pm LaF3 (11) UCl3 (9) PuBr3 (8) PuBr3 (8)Sm YF3 (9) UCl3 (9) PuBr3 (8) FeCl3 (6)Eu YF3 (9) UCl3 (9) PuBr3 (8)Gd YF3 (9) UCl3 (9) FeCl3 (6) FeCl3 (6)Tb YF3 (9) PuBr3 (8) FeCl3 (6) FeCl3 (6)Dy YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Ho YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Er YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Tm YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Yb YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Lu YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)Y YF3 (9) AlCl3 (6) FeCl3 (6) FeCl3 (6)

Values in parentheses indicate the coordination number.

SPH SPH


Binary Compounds 27

Figure 3.3The LaF3 structure (reproduced by permission of Macmillan from S.A. Cotton, Lanthanides andActinides, 1991, p. 42).

Figure 3.4The structure of UCl3 and certain LnCl3 [reproduced in modified form, from R.B. King (ed.),Encyclopedia of Inorganic Chemistry, 1st edition, Wiley, Chichester; 1994].

trigonal prism (Figure 3.4); if one of the face-capping halogens is removed from the ‘UCl3’structure, the 8 coordinate PuBr3 structure is generated. Finally, the AlCl3 and BiI3 (alsosometimes referred to as ‘FeCl3’) structures both have octahedral six-coordination.

The trends are similar to those found in halides of the Group I and Group II metalsand are explained on an ionic packing model; more anions can be packed round the largelanthanide ions early in the series than around the smaller, later ones; similarly, more ofthe small fluoride ions can be packed round a given lanthanide ion than is the case with themuch larger iodide anion.

Properties of the Trihalides

The trihalides are high-melting solids, with many uses in synthetic chemistry, though theirinsolubility in some organic solvents means that complexes, such as those with thf (Section4.3.3), are often preferred.

3.3.2 Tetrahalides

Only the fluorides of Ce, Pr, and Tb exist, the three lanthanides with the most stable (+4)oxidation state. Fluorine is most likely to support a high oxidation state, and even thoughsalts of ions like [CeCl6]2− are known, the binary chloride has not been made. CeF4 can becrystallized from aqueous solution as a monohydrate. Anhydrous LnF4 (Ln = Ce, Pr, Tb)

SPH SPH



can be made by fluorination of the trifluoride or, in the case of Ce, by fluorination ofmetallic Ce or CeCl3. All three tetrafluorides have the MF4 structure with dodecahedraleight coordination. Factors that favour formation of a tetrafluoride include a low value ofI4 for the metal and a high lattice energy (see Section 2.9.1). This is most likely to be foundwith the smallest halide ion, fluoride. The low bond energy of F2 is also a supporting factor.

3.3.3 Dihalides

These are most common for metals with a stable (+2) state, such as Eu, Yb, and Sm.As would be expected, they most often occur for the iodide ion, the best reducing agent.A number of dihalides are known for other metals, though some of these do not actuallyinvolve the (+2) oxidation state.

Synthetic Routes

These compounds are usually made by reduction using hydrogen (e.g. EuX2, YbX2, orSmI2) or reproportionation.

2 EuCl3 + H2 → 2 EuCl2 + 2 HCl

2 DyCl3 + Dy → 3 DyCl2

In a few cases, thermal decomposition is applicable [LnI2 (Ln = Sm, Yb); EuBr2].

2 YbI3 → 2 YbI2 + I2

Another method, used especially for diiodides, involves heating the metal with HgX2.

Tm + HgI2 → TmI2 + Hg

The known dihalides are listed in Table 3.2, together with an indication of the structure andcoordination number.

The iodides of Nd, Sm, Eu, Dy, Tm, and Yb are definitely compounds of the +2 ions, withsalt-like properties, are insulators, and have magnetic and spectroscopic properties expected

Table 3.2 Structures of the lanthanide dihalides

F Cl Br I

La aMoSi2 (8)Ce aMoSi2 (8)Pr aMoSi2 (8)Nd PbCl2 (9) PbCl2 (9) SrBr2 (7,8)PmSm CaF2 (8) PbCl2 (9) PbCl2 (9); SrBr2 (7,8) EuI2 (7)Eu CaF2 (8) PbCl2 (9) SrBr2 (7,8) EuI2 (7)Gd aMoSi2 (8)TbDy SrBr2 (7,8) SrI2 (7) CdCl2 (6)HoErTm SrI2 (7) SrI2 (7) CdI2 (6)Yb CaF2 (8) SrI2 (7) SrI2 (7); CaCl2 (6) CdI2 (6)Lu

a LnIII compounds with the structure M3+(I−)2(e−).

SPH SPH


Binary Compounds 29

for the M2+ ions (thus electronic spectra of EuX2 resemble those of the isoelectronic Gd3+

ions). SmI2 is proving an important reagent in synthetic organic chemistry (Section 8.3).Several of the diiodides, however, have a metallic sheen and are very good conductors of

electricity, such as LaI2, CeI2, PrI2, and GdI2. Since they are good conductors in the solidstate, the presence of delocalized electrons is indicated. M3+(I−)2(e−) is a likely structure.None of these metals exhibits a stable +2 state in any of its compounds and consideration offactors such as their ionization energies suggest they are unlikely to form stable compoundsin this state (see Section 2.9.2).

3.3.4 Oxides

Oxides M2O3

As already mentioned, the metals burn easily, rather like Group 2 metals, forming oxides(but with some nitride). The oxides are therefore best made by thermal decomposition ofcompounds like the nitrate or carbonate.

4 Ln(NO3)3 → 2 Ln2O3 + 12 NO2 + 3 O2

Most lanthanides form Ln2O3, but those metals with accessible +4 and +2 oxidation statescan afford other stoichiometries. These may be turned into Ln2O3 by synthesis under areducing atmosphere. CeO2 can be reduced to Ce2O3 using hydrogen. The oxides are some-what basic and absorb CO2 from the atmosphere, forming carbonates, and water vapour,forming hydroxides. As expected they dissolve in acid, forming salts, and are convenientstarting materials for the synthesis of lanthanide salts, including the hydrated halides.

The sequioxides Ln2O3 adopt three structures, depending upon the temperature andupon the lanthanide involved. At room temperature, La2O3 to Sm2O3 inclusive adopt theA-type structure, which has capped octahedral 7 coordination of the lanthanide. The B-typestructure tends to be exhibited by La2O3 to Sm2O3 at higher temperatures and has threedifferent lanthanide sites, one with distorted 6 coordination and the others with face-cappedoctahedral 7 coordination. The C-type structure is followed by heavier metals and has 6coordination, severely distorted from an octahedron.

Oxides MO2

Cerium, having the stablest (+4) state, is the only metal to form a stoichiometric oxide inthis state, CeO2, which has the fluorite structure. It is white when pure, but even slightlyimpure specimens tend to be yellow. It can be made by burning cerium or heating saltslike cerium(III) nitrate strongly in air. It is basic and dissolves with some difficulty in acid,forming Ce4+(aq), which can be isolated as salts such as the nitrate Ce(NO3)4.5H2O. Usesof CeO2 include self-cleaning ovens and as an oxidation catalyst in catalytic converters.

Praseodymium and terbium form higher oxides, of which a number of phases are knownbetween Ln2O3 and LnO2. Ignition of praseodymium nitrate leads to Pr2O3 but furtherheating in an oxygen atmosphere gives Pr6O11 or even PrO2; terbium similarly yieldsTb4O7 and TbO2.

Oxides MO

Reduction of Eu2O3 with Eu above 800 ◦C (comproportionation) gives EuO.

Eu2O3 + Eu → 3 EuO

SPH SPH



This compound and the similar YbO have salt-like (NaCl) structures and are genuine LnII

compounds, being insulators. Similar comproportionation methods using high pressureshave been used to obtain SmO and NdO; these are shiny conducting solids, probablycontaining Ln3+ ions.

Oxide Superconductors

Superconductivity is the phenomenon in which a material conducts electricity with virtu-ally zero resistance. For many years until the mid 1980s, the highest temperatures availablewere around 20 K and required expensive liquid helium (or hydrogen) coolant. If high-temperature superconductors can be made, this has obvious application in area such aspower transmission. In 1986, Bednorz and Muller reported that La1.8Sr0.2CuO4 was a su-perconductor up to 38 K. This sparked intense world-wide activity, and the following yearWu, Chu and others reported that YBa2Cu3O7−δ (0 ≤ δ ≤ 1) had a Tc (superconductingtransition temperature) of 92 K, bringing the phenomenon into the liquid nitrogen range.Steady though less spectacular advances have produced materials like HgBa2Ca2Cu3O8

(Tc = 133K).The structure of YBa2Cu3O7 is based on an oxygen-deficient layered perovskite structure

and features two types of copper environment (Figure. 3.5), formally containing Cu2+ andCu3+, with both square planar and square pyramidal coordination. Removing the shadedoxygens results in phases down to the semiconductor YBa2Cu3O6. The mechanism ofsuperconductivity is still debated, but one theory suggests that it involves an electron passingthrough the lattice distorting it in such a way that a second electron follows closely in its wake

Figure 3.5The structure of YBa2Cu3O7; the shaded atoms are those removed to create oxygen-deficient phasesup to YBa2Cu3O6 (reproduced by permission of Macmillan from S.A. Cotton, Lanthanides andActinides, 1991, p. 47).

SPH SPH


Hydrides 31

with no hindrance to its passing, the electron pair being known as a ‘Cooper pair’. Despiteintense activity over the last 15 years no truly commercial material has yet emerged, due tothe intrinsically brittle nature of the oxide ceramic materials making fabrication into wiresimpossible. Current thinking favours making thin films, though with present technologythey will only have low current-carrying capacity.

3.4 Borides

A number of stoichiometries obtain, such as LnB2, LnB4, LnB6, LnB12, and LnB66.The most important are LnB6. Borides are obtained by heating the elements together

at 2000 ◦C or by heating the lanthanide oxide with boron or born carbide at 1800 ◦C.They are extremely unreactive towards acids, alkalis, and other chemicals, and are metallicconductors. They contain a continuous three-dimensional framework of (B6)2− octahedrainterspersed with Ln3+ ions, indicating an electronic structure (Ln3+ ) (B6)2− (e− ) in mostcases, though EuB6and YbB6 are insulators, suggesting them to be (Ln2+ ) (B6)2−. Becauseof its high thermal stability and melting point (2400 ◦C), and its metal-like electrical andthermal conductivity, LaB6 is an important thermionic emitter material for the cathodes ofelectron guns. Mixed boride materials are also of commercial importance. Nd–Fe–B alloys,with compositions such as Nd2Fe14B, are the strongest permanent magnet materials avail-able today. Other important compounds include lanthanide rhodium boride low-temperaturesuperconductors (such as ErRh4B4), part of a wider LnM4B4 family (M = transition metal).

3.5 Carbides

The lanthanides form carbides with a range of compositions, notably LnC2, but additionallyLn2C3, LnC, Ln2C, and Ln3C phases are known (depending upon the lanthanide in questionand the conditions of synthesis). LnC2 adopt the CaC2 structure, containing isolated C2−

2ions.

3.6 Nitrides

These can be made by direct synthesis from the elements at 1000 ◦C. They have the NaClstructure and can be hydrolysed to NH3.

3.7 Hydrides

Ternary hydrides such as LaNi5H6 have attracted attention as materials for electrodes in fuelcells and for gas storage. Nickel/metal hydride batteries for notebook PCs are a less toxic al-ternative than Ni/Cd batteries; when charging, hydrogen generated at the negative electrodeenters the lattice of the La/Ni alloy (in practice a material such as La0.8Nd0.2Ni2.5Co2.4Si0.1

is used to improve corrosion resistance, storage capacity, discharge rate, etc.). The overallcell reaction is:

LaNi5 + 6 Ni(OH)2 � LaNi5H6 + 6 NiOOH

SPH SPH



The lanthanides also form simple binary hydrides on combination of the elements at about300 ◦C. These compounds have ideal compositions of MH2 and MH3, but are frequentlynon-stoichiometric. Thus lutetium forms phases with ranges LuH1.83 to LuH2.23 and LuH2.78

to LuH3.00. Reaction of ytterbium with hydrogen under pressure gives YbH2.67. MH3 areobtained only at higher gas pressures, whilst europium, the lanthanide with the most stable+2 state, forms only EuH2. The dihydrides are generally good electrical conductors, andthus are thought to be M3+(H−)2(e−), whilst the trihydrides are salt-like nonconductorsbelieved to be M3+(H−)3. The hydrides are reactive solids, owing to the presence of theeasily hydrolysed H− ion.

3.8 Sulfides

These are quite important compounds. A number of stoichiometries exist, the most importantbeing Ln2S3. These can be made by direct synthesis, heating the elements together, or bypassing H2S over heated LnCl3. Eu2S3 cannot be prepared by this latter route.

A range of compositions between Ln2S3 and Ln3S4, the latter formed when Ln2S3 losesulfur on heating, generally occurs. Ln2S3 are insulators, genuine Ln(III) compounds, butLn3S4(having the Th3P4 structure) are more complex. Some, like Ce3S4, are metallic con-ductors and are thus (Ln3+)3(S2−)4(e−); others, like Eu3S4 and Sm3S4, are semiconductorsand may be (Ln2+)(Ln3+)2(S2−)4. The structures of Ln2S3 fall into a pattern. La2S3 to Dy2S3

adopt the ‘Gd2S3’ structure with 7-coordinate lanthanide ions; Dy2S3 to Tm2S3 have the‘Ho2S3’ structure with 6- and 7-coordinate lanthanides, whilst Yb2S3 and Lu2S3 have thecorundum structure with just 6 coordination.

Monsulfides MS are formed by direct combination. They adopt the NaCl structure buthave a variety of bonding types. YbS and EuS are genuine Ln2+ S2− but CeS exhibits themagnetic properties expected for Ce3+ and has a bronze metallic lustre as well, so it thoughtto be (Ce3+)(S2−)(e−). This substance not only has electrical conductivity in the metallicregion but also can be machined like a metal too. SmS has some unusual properties; it isusually obtained as a black semiconducting phase, Sm2+ S2−, but can reportedly be turnedinto a golden metallic phase by the action of pressure, polishing or even scratching on singlecrystals.

Oxysulfides are also rather important. Y2O2S is used as a host material for Ln3+ ionemitters (e.g. Eu, Tb) in some phosphors, notably those used in TV screens. When mis-chmetal is used to remove oxygen and sulfur from impure iron and steel, the product isan oxysulfide, which forms an immiscible solid even in contact with molten steel and thusdoes not contaminate the product.

Question 3.1. Why do Eu and Yb have lower boiling points than the other lanthanides andsignificantly higher atomic radii?Answer 3.1 Most lanthanides can be described by an electronic structure (Ln3+) (e−)3,whereas metals like Eu and Yb can be better described by (Ln2+) (e−)2. The greater chargeon the (+3) ions draws electrons in closer, so it is expected that the metals with this electronicstructure will have smaller atoms. Similarly the greater number of electrons in the (Ln3+)(e−)3 structure leads to stronger metallic bonding than in metals with the structure (Ln2+)(e−)2 and hence the weaker cohesive force requires less energy to overcome it, resulting inlower enthalpies of vapourization and lower boiling points.

SPH SPH


Sulfides 33

Question 3.2 Suggest why nitrogen is not a suitable inert atmosphere for the followingreaction:

2 LnF3 + 3 Ca → 2 Ln + 3 CaF2

Answer 3.2 Like magnesium, which forms Mg3N2, hot lanthanides react with nitrogen,forming the nitride LnN (see Section 3.9).

Question 3.3 Produce a balanced equation for the reaction of LaCl3 with silica.Answer 3.3

2 LaCl3 + SiO2 → 2 LaOCl + SiCl4

Question 3.4 Studying the data in Table 3.1 if necessary, explain the patterns in coordinationnumber of the lanthanide trihalides.Answer 3.4 As the radius of the lanthanide ion decreases on crossing the series from leftto right, the coordination number of the metal in a given halide decreases; a similar effectis seen on descending Group 7(17), as the radius of the halide ion increases.

Question 3.5 The melting and boiling points of LaX3 (X = Cl, Br, I) are 862/1750 ◦C(X = Cl); 789 and 1580 ◦C (X = Br); 778 and 1405 ◦C (X = I). Comment on these values.Answer 3.5 The melting point is an indication of the magnitude of the lattice energy,whilst the boiling point is a measure of the attraction between the mobile ions. Latticeenergy decreases as the size of the anions increase, so it would be least for the iodide.Covalent character in the bonding, which would also decrease the melting point, increasesin the same direction.

Question 3.6 Why could CeO2 be a useful catalyst for oxidations?Answer 3.6 As with V2O5 in the Contact process for making sulfuric acid, the ability ofcerium to adopt transition-metal-like behaviour in switching oxidation states means that thecerium oxide can effectively act as an oxygen-storage system.

Question 3.7 LnC2 are good conductors of electricity and form ethyne, C2H2, on hydrolysis(as does CaC2, traditionally used in ‘acetylene’ headlamps on early cars and bicycles).Explain these observations.Answer 3.7 These compounds have the structure M3+( C2−

2 )(e−). The delocalized electronscause the high conductivity. The ethynide ions react with water thus:

C2−2 (s) + 2 H2O(1) → C2H2(g) + 2 OH−(aq)

Question 3.8 Suggest an equation for the reaction of LaCl3 with H2S.Answer 3.8

2 LaCl3 + 3 H2S → La2S3 + 6 HCl

SPH SPH


34

...


4 Coordination Chemistryof the Lanthanides


� recall which donor atoms and types of ligand form the most stable complexes;� explain why later lanthanides form complexes with greater stability constants than do the

earlier lanthanides;� explain why polydentate ligands form more stable complexes;� recall and explain the most common coordination numbers in lanthanide complexes;� recall the common geometries of complexes;� work out coordination numbers and suggest geometries of complexes, given a formula;� explain the choice of particular compounds as Magnetic Resonance Imaging agents.

4.1 Introduction

Forty years ago, very little was known about lanthanide complexes. By analogy with thed-block metals, it was often assumed that lanthanides were generally six coordinate in theircomplexes. We now know that this is not the case, that lanthanides (and actinides) show awider variety of coordination number than do the d-block metals, and also understand thereasons for their preferred choice of ligand.

4.2 Stability of Complexes

For the reaction between a metal ion and a ligand

Mn+ + Ly− ⇔ ML(n−y)+

a stability constant may be defined approximately

β1 = [ML(n−y)+]/[Mn+][Ly−]

Table 4.1 lists log[β1] values for La3+ and Lu3+as well as Sc3+ and Y3+with a numberof common ligands, whilst Table 4.2 gives the log[β1] values for all the lanthanide ionscomplexing with EDTA4−, DTPA5−, and fluoride.

The values for Lu3+ are greater than those for La3+ as the smaller Lu3+ ion has a greatercharge density and stronger electrostatic attraction for a ligand. The stability constants forthe smaller halide ligands are greater on account of their higher charge density. Values formultidentate ligands are greater because of entropy factors (see below) and also because


...


36 Co-ordination Chemistry of the Lanthanides

Table 4.1 Aqueous stability constants (log β1) for lanthanide (3+) and other ions

Ligand I (mol/dm3) La3+ Lu3+ Y3+ Sc3+ Fe3+ Cu2+ Ca2+ U4+ UO2+2 Th4+

F− 1.0 2.67 3.61 3.60 6.2a 5.2 0.9 0.6 7.78 4.54 7.46Cl− 1.0 −0.1 −0.4 −0.1 0 0.63 −0.06 −0.11 0.30 −0.10 0.18Br− 1.0 −0.2 −0.15 −0.07 −0.2 −0.5 0.2 −0.3 −0.13NO−

3 1.0 0.1 −0.2 0.3 −0.5 −0.01 −0.06 0.3 −0.3 0.67a

OH− 0.5 4.7 5.8 5.4 9.0 11.27 6.3a 1.0 12.2 8.0a 9.6a

acac− 0.1 4.94 6.15 5.89 8 10 8.16 8EDTA4− 0.1 15.46 19.8 18.1 23.1 25.0 18.7 10.6 25.7 7.4 25.3DTPA5− 0.1 19.5 22.4 22.4 24.4 28 21.4 10.8 28.8OAc− 0.1 1.82 1.85 1.68 3.38 1.83 0.5a 2.61 3.89glycine 0.1 3.1 3.9 3.5 10.0a 8.12 1.05

a = Data for solutions of slightly different ionic strength

Table 4.2 Aqueous stability constants (log β1) for lanthanide (3+) ions with fluoride, EDTA, and DTPA

Ligand Y3+ La3+ Ce3+ Pr3+ Nd3+ Pm3+ Sm3+ Eu3+ Gd3+ Tb3+ Dy3+ Ho3+ Er3+ Tm3+ Yb3+ Lu3+

F− 3.60 2.67 2.87 3.01 3.09 3.16 3.12 3.19 3.31 3.42 3.46 3.52 3.54 3.56 3.58 3.61EDTA4− 18.08 15.46 15.94 16.36 16.56 17.10 17.32 17.35 17.92 18.28 18.60 18.83 19.30 19.48 19.80DTPA5− 22.05 19.48 20.33 21.07 21.60 22.34 22.39 22.46 22.71 22.82 22.78 22.74 22.72 22.62 22.44

once one end of a ligand is attached, there is a higher chance of the other donor atomsattaching themselves (conversely, with a multidentate ligand, if a bond to one donor atomis broken, the others hold).

In a graph of log K against Z (the atomic number), there is a smooth increase expectedas the ionic radius decreases as the greater charge density of the smaller ions leads to morestable complexes. There are inflections, particularly around Gd, possibly due to the changein coordination number of the aqua ion.

For these complexation reactions, �H has small values, either exothermic or endothermic;the main driving force for complex formation, particularly where multidentate ligands areinvolved, is the large positive entropy change. Thus for:

[La(OH2)9]3+(aq) + EDTA4−(aq) → [La(EDTA)(OH2)3]−(aq) + 6 H2O(l)

�G = −87 kJ mol−1; �H = −12 kJ mol−1; �S = +251 JK−1mol−1

In the reaction

LnX3.9H2O(s) + 3 L−(aq) → [Ln(L)3]3+(aq) + 3X− + 9 H2O(l)

where L = 2,6-dipicolinate and X is bromate or ethyl sulfate, there is a smooth variation in�H across the series, as there is no change in the coordination number of the aqua ion.

In general, lanthanide ions prefer to bind to hard donors such as O and F, rather than to softbases such as P and S donor ligands. Nitrogen-containing ligands are an apparent anomaly,since they form relatively few complexes; this is at least partly due to the high basicity ofsuch ligands, tending to result in the precipitation of hydroxides, etc. (Lanthanide ammineshave been made in solvents like supercritical NH3.) The higher electronegativity of oxygenand consequent polar nature of ligands like phosphine oxides R3Pδ+==Oδ− is also a factor.Use of nonaqueous solvents of weak donor ability (e.g. MeCN) can lead to the isolation ofcomplexes decomposed by water.

...


Complexes 37

4.3 Complexes

4.3.1 The Aqua Ions

The coordination number of [Ln(H2O)n]3+is believed to be 9 for the early lanthanides(La–Eu) and 8 for the later metals (Dy–Lu), with the intermediate metals exhibiting amixture of species. The nine-coordinate species are assigned tricapped trigonal prismaticstructures (Figure 4.1) and the eight-coordinate species square antiprismatic coordination.A considerable amount of spectroscopic data has led to this conclusion; for example, thevisible spectrum of Nd3+(aq) and [Nd(H2O)9]3+ ions in solid neodymium bromate arevery similar to each other and quite different to those of Nd3+ ions in eight-coordinateenvironments, correlating with solution X-ray and neutron-diffraction studies, indicatingn ∼ 8.9 in solution. Similarly, neutron-diffraction studies indicate n ∼ 8.5 for Sm3+(aq)and 7.9 for Dy3+(aq) and Lu3+(aq), whilst solution luminescence studies suggest n valuesof 9.0, 9.1, 8.3, and 8.4 for Sm3+, Eu3+, Tb3+(aq), and Dy3+ respectively.

4.3.2 Hydrated Salts

These are readily prepared by reaction of the lanthanide oxide or carbonate with the acid.Salts of noncoordinating anions most often crystallize as salts [Ln(OH2)9]X3 (X e.g. bro-mate, triflate, ethyl sulfate, tosylate). These contain the [Ln(OH2)9]3+ ion (Figure 4.1),even for the later lanthanides, where in aqueous solution the eight-coordinate species[Ln(OH2)8]3+ predominates.

The lanthanide–water distances for the positions capping the prism faces and at thevertices are different; on crossing the series from La to Lu, the Ln–O distance decreasesfrom 2.62 to 2.50 A for the three face-capping oxygens but change more steeply from2.52 to 2.29 A for the six apical oxygens (data for the triflate). In contrast, the perchlo-rate salts are [Ln(OH2)6](ClO4)3 and eight-coordinate species are found in species like[Er(OH2)8](ClO4)3.(dioxane).2H2O, [Eu(OH2)8]2(V10O28).8H2O, and as [Gd(OH2)8]3+

ions encapsulated inside a crown ether ring. Many of these compounds have lattices heldtogether by hydrogen bonds, a factor which clearly affects the solubility and hence thecompound isolated.

When anions can coordinate, a wide variety of species obtains. Some examples follow.Among the nitrates, [Ln(NO3)3.(H2O)5] (Ln = La, Ce) have 11-coordinate lanthanides

Figure 4.1The structure of the nonaaqualanthanide ion (reproduced by permission of the American ChemicalSociety from B.P. Hay, Inorg. Chem., 1991, 30, 2881).

...



Figure 4.2Structure of the dimeric cation [(H2O)7LnCl2Ln(H2O)7]4+ (reproduced by permission of the Interna-tional Union of Crystallography from A. Habenschuss and F.H. Spedding, Cryst. Struct. Commun.,1982, 7, 538).

whilst in [Ln(NO3)3.(H2O)4] (Ln = Pr–Yb, Y) 10 coordination is the rule, and two hydratescontaining 9-coordinate [Lu(NO3)3.(H2O)3] species have been characterized; in general, inlanthanide complexes a coordinated nitrate group is almost always bidentate. The chloridesand bromides of La and Ce, LnX3. 7H2O, are dimeric [(H2O)7Ln(µ-X)2Ln(OH2)7]X4 (X =Cl, Br; Figure 4.2), whilst LnCl3.6H2O (Ln = Nd–Lu) have antiprismatic [LnCl2(OH2)6]+

ions.Amongst the hydrated bromides, LnBr3.6H2O (Ln = Pr–Dy) are [LnBr2(OH2)6]Br, like

the heavier rare earth chlorides; and LnBr3.8H2O (Ln = Ho–Lu) are [Ln(OH2)8]Br3. Thestructures of the iodides LnI3.9H2O and LnI3.8H2O also involve pure aqua ions. Severalhydrated sulfates exist, where both water and sulfates are bound to the metal, usually with 9coordination. Thiocyanates exist as Ln(NCS)3(H2O)6(Ln = La–Dy) and Ln(NCS)3(H2O)5

(Ln = Sm–Eu) molecules.The acetates, unlike the acetates of transition metals like CrIII, FeIII, and RuIII, do not

adopt oxo-centred structures with M3O cores. Instead, [Ln(OAc)3.1.5(H2O)] (Ln = La–Pr)have structures with acetate-bridged chains crosslinked by further acetate bridges;[Ln(OAc)3.(H2O)] (Ln = Ce–Pr) have one-dimensional polymeric structures with acetatebridges; and [Ln(OAc)3.4H2O)] (Ln = Sm–Lu) are acetate-bridged dimers.

Adding oxalate ions to a solution of lanthanide ions affords quantitative precipitation ofthe oxalate; this is used in traditional quantitative analysis of lanthanides, as on ignitionthey are converted into the oxide, the weighing form.

The chloride, bromide, bromate, perchlorate, nitrate, acetate, and iodide salts are allsoluble in water, the sulfates sparingly soluble, and the fluorides, carbonates, oxalates, andphosphates insoluble.

4.3.3 Other O-Donors

Complexes of lanthanide salts with neutral donors are generally made by mixing solutions ofthe ligand and the metal salt dissolved in a nonaqueous solvent such as ethanol or acetonitrile.Complexes of ligands like phosphine and arsine oxides were the first complexes of this typeto be studied in detail. The best defined series involve hexamethylphosphoramide [hmpa:(Me2N)3PO] as a ligand; two series are obtained in which the stoichiometry is maintainedacross the series, Ln(hmpa)6(ClO4)3 and Ln(hmpa)3Cl3, with structural confirmation for

...


Complexes 39

[Nd(hmpa)6](ClO4)3 and mer-Ln(hmpa)3Cl3 (Ln = Pr, Dy, Yb). Apart from the usualsynthesis involving the ligand and the appropriate metal salt in ethanol, such complexescan also be made by an unusual ‘one-pot’ synthesis leading to the isolation of 9-coordinateLa(hmpa)3(NO3)3, 7-coordinate La(hmpa)4(NCS)3 and La(hmpa)4Br3.

LnNH4X/HMPA−−−−−−−−−→toluene, heat

Ln(hmpa)nX3 (X = NO3, n = 3; X = Br, NCS, n = 4)

Complexes of Ph3PO and Ph3AsO, though well known, have yet to be fully studied. Thebulky phenyl groups exert at-a-distance steric effects, yet allow several ligands of this typeto coordinate to a metal ion, as in [La(Ph3PO)5Cl] (FeCl4)2 and [Ln(Ph3PO)4Cl2] (CuCl3),both of which have six-coordinate lanthanides. Early lanthanides form Ln(Ph3PO)4(NO3)3

(La–Nd), which have two bidentate and one monodentate nitrates, giving 9-coordinate lan-thanides; Lu(Ph3PO)4(NO3)3 is actually [Ln(Ph3PO)4(NO3)2] NO3, with 8 coordination.Nine coordination is also known to occur in Eu(Ph3AsO)3(NO3)3. On the other hand, witha less bulky ligand, lanthanum attains 10 coordination in La(Ph2MePO)4(NO3)3. ‘Solvent’is coordinated in Ln(Ph3PO)2(NO3)3(EtOH) (Ln = Ce–Eu), also 9-coordinate. TriflatesLn(Ph3PO)4(OTf)3(OTf = CF3SO3) contain [Ln(Ph3PO)4(OTf)2]+ ions; for La and Nd,one triflate is bidentate and one monodentate, whilst for Er–Lu both triflates are monoden-tate. A few halide complexes have been characterized structurally, notably for Y, including[Y(Me3PO)6] Br3 and [Y(Ph3PO)4Cl2] Cl, though it would be expected that lanthanidecomplexes would be analogous.

Complexes of tetrahydrofuran (thf) have been studied in considerable detail in recentyears. The anhydrous metal chlorides, widely used in synthesis, are relatively insoluble inorganic solvents whilst the ease of replacement of the labile thf molecules in the relativelysoluble thf complexes makes them more useful synthetic reagents. Several types of complexhave been characterized, including [LaCl3(thf)4] (seven coordinate); LnCl3(thf)3.5, actu-ally [LnCl2(thf)5]+ [LnCl4(thf)2]−, containing both seven- and six-coordinate lanthanides(formed by metals such as Sm and Eu); and octahedral [LnCl3(thf)3], only found for Yband Lu (see Question 4.3, below).

Other complexes to have been studied include Nd(thf)4Br3 and Yb(thf)4(NCS)3, bothpentagonal bipyramidal seven-coordinate molecules with two axial halides (generallyadopted for LnL4X3 systems). The thiocyanate complexes of the earlier lanthanides,[Ln(thf)4(NCS)3]2, are dimers with two bridging thiocyanates affording eight coordina-tion. Nitrate complexes Pr(thf)4(NO3)3 and Yb(thf)3(NO3)3 have been characterized, asrecently have the iodides LnI3(THF)4 [Ln = Pr] and LnI3(THF)3.5 [Ln = Nd, Gd, Y], thelatter being [LnI2(THF)5][LnI4(THF)2], analogous to some of the chlorides.

Another example of the lanthanide contraction at work, verified by crystallographers,lies in the complexes of the lanthanide nitrates with dimethyl sulfoxide (dmso). The earlylanthanides form 10-coordinate Ln(dmso)4(NO3)3 (Ln = La–Sm), whilst the heavier metalsform Ln(dmso)3(NO3)3(Ln = Eu–Lu, Y).

4.3.4 Complexes of β-Diketonates

These are an important class of compounds with a general formula Ln(R1.CO.CH.CO.R2)3.The acetylacetonates, Ln(acac)3 (R1 = R2 = CH3) can readily be made from a lanthanidesalt and acetylactone by adding sodium hydroxide:

LnX3 + 3 Na(acac) → Ln(acac)3 + 3 NaX

...



(a)

(b)

Figure 4.3(a) The structure of Ho(PhCOCHCOPh)3(H2O) (reproduced by permission of the American ChemicalSociety from A. Zalkin, D.H. Templeton, and D.G. Karraker, Inorg. Chem., 1969, 8, 2680). (b) Thestructure of La(acac)3(H2O)2 (reproduced by permission of the American Chemical Society from T.Phillips, D.E. Sands, and W.F. Wagner, Inorg. Chem., 1968, 7, 2299).

They crystallize as hydrates Ln(acac)3.(H2O)2(Ln = La–Ho, Y) and Ln(acac)3.(H2O)(Ln e.g. Yb) though there is a tendency to contain lattice water as well (Figure 4.3 foracac and PhCOCHCOPh complexes). These compounds are difficult to dehydrate, evenin vacuo; they decompose on heating, and on dehydration at room temperature tend tooligomerize to involatile materials. They form Lewis base adducts like Ln(acac)3.(Ph3PO)and Ln(acac)3.(phen) (7- and 8-coordinate, respectively).

Using bulkier R groups (R1 = R2 = CMe3, for example) affords more congested diketo-nate complexes that are more tractable; Ln(Me3.CO.CH.CO.CMe3)3 [often abbreviated toLn(dpm)3 or Ln(tmhd)3] are monomers in solution and sublime in vacuo at 100–200 ◦C; inthe solid state they are dimers for Ln = La–Dy (CN 7) and monomers for Dy–Lu (CN 6)and have trigonal prismatic coordination. They tend to hydrate readily, forming adducts[e.g. capped trigonal prismatic Ln(dpm)3(H2O)].

Complexes of fluorinated diketones (R1 = CF3, R2 = CH3, L = tfac; R1 = R2 = CF3, L =hfac; R1 = cyclo-C4H3S, R2 = CF3, L = tta; R1 = CF3CF2CF2, R2 = CMe3, L = fod) arealso important. Again, the initial complexes obtained in synthesis are hydrates that can bedehydrated in vacuo. 2-Thenoyltrifluoroacetone complexes Ln(tta)3.2H2O are important in

...


Complexes 41

solvent extraction; addition of phosphine oxides gives a synergistic improvement in extrac-tion owing to the formation of phosphine oxide complexes. Phosphine oxide complexes ofhfac, Ln(hfac)3(Bu3PO)2, can be separated by gas chromatography.

4.3.5 Lewis Base Adducts of β-Diketonate Complexes

Because the diketonates Ln(R1.CO.CH.CO.R2)3 are co-ordinatively unsaturated, they com-plete their coordination sphere by forming adducts with Lewis bases, as already noted.Because of the paramagnetism of nearly all of the lanthanide ions, there are resulting shiftsin the resonances in the NMR spectrum of any organic molecule coordinated and thismeans that these complexes were formerly important as NMR Shift Reagents (see Section5.5.1). Their stoichiometry and structure was consequently well researched. Both 1:1 and1:2 adducts are possible, depending on steric factors of both the complex and the ligand.Solubility factors may also mean that the most abundant species in a solution is not thatwhich crystallizes. Ln(fod)3 forms both 1:1 and 1:2 complexes with even quite bulky bases(e.g. 2,4,6-Me3py). In contrast, the more bulky Eu(dpm)3 forms solid Eu(dpm)3L2 adductswith both pyridine and C3H7NH2, but only Eu(dpm)3L with the bulkier 2-methylpyridine.Detailed studies of Ho(dpm)3 have shown evidence only for 1:1 complexes with stericallydemanding ligands like Ph3PO, camphor, and borneol.

The structures are known for a number of these compounds; thus, of the eight-coordinate compounds, Eu(acac)3(phen) is square antiprismatic, Eu(dpm)3(py)2 andNd(tta)3(Ph3PO)2 are dodecahedral; seven-coordinate Lu(dpm)3(3-picoline) is a face-capped trigonal prism and Eu(dpm)3(quinuclidine) a capped octahedron. Nine-coordinationis found in Eu(dpm)3(terpy) and also in Pr(facam)3(DMF)3(facam = 3-trifluoroacetyl-D-camphorate).

Anionic species of the type [Ln(R1.CO.CH.CO.R2)4]− are another means of at-taining coordinative saturation and are well known, such as the antiprismatic[Eu(Ph.CO.CH.CO.Ph)4]− ion.

4.3.6 Nitrate Complexes

These exhibit high coordination numbers. Double nitrates Mg3Ln2(NO3)12.24H2O, whichcontain [Ln(NO3)6]3− ions, are important historically as they were once used for the sepa-ration of the lanthanides by fractional crystallization. Use of counter-ions like Ph4P+ andMe4N+ allow isolation of the 10-coordinate [Ln(NO3)5]2− ions.

4.3.7 Crown Ether Complexes

In contrast to the alkali metals, the lanthanides do not form crown ether complexes readilyin aqueous solution, due to the considerable hydration energy of the Ln3+ ion. These com-plexes are, however, readily synthesized by operating in non-aqueous solvents. Becausemany studies have been made with lanthanide nitrate complexes, coordination numbersare often high. Thus 12-coordination is found in La(NO3)3(18-crown-6) (Figure 4.4), 11coordination in La(NO3)3(15-crown-5), and 10 coordination is found in La(NO3)3(12-crown-4). Other complexes isolated include Nd(18-crown-6)0.75(NO3)3, which is in fact[{Nd(18-crown-6)(NO3)2}+]3 [Nd(NO3)6]. Other lanthanide salts complex with crownethers; small crowns like 12-crown-4 give 2:1 complexes with lanthanide perchlorates,though the 2:1 complexes are not obtained with lanthanide nitrates where the anion can

...



Figure 4.4The structure of La(NO3)3(18-crown-6) (reproduced by permission of the Royal Society of Chemistryfrom J.D.J. Backer-Dirks, J.E. Cooke, A.M.R. Galas, J.S. Ghotra, C.J. Gray, F.A. Hart, and M.B.Hursthouse, J. Chem. Soc., Dalton Trans., 1980, 2191).

readily coordinate. Lanthanide chloride complexes often include water in the co-ordinationsphere; thus the complex with the formula ErCl3(12-crown-4).5H2O actually contains 9-coordinate [Er(12-crown-4)(H2O)5]3+ cations whilst NdCl3(18-crown-6).4H2O is [Nd(18-crown-6)Cl2(H2O)2]+(Cl−).2H2O. On reaction in MeCN/MeOH, 15-crown-5 reacts withneodymium chloride to form complexes with [Nd(OH2)9]3+ and [NdCl2(OH2)6]+ ionshydrogen-bonded to the crown ether without any direct Nd–crown ether bonds. By carryingout electrocrystallization, the water-free [Nd(15-crown-5)Cl3] is obtained. Lanthanides alsocomplex with the noncyclic linear polyethers such as glyme and with other macrocyclessuch as the calixarenes.

4.3.8 Complexes of EDTA and Related Ligands

These complexes are not only interesting on account of their chemical and structural prop-erties, but they have an important place in the history of lanthanide separation chemistry.Although lanthanide ions do not form strong complexes with N-donor ligands, EDTA andother polyaminopolycarboxylic acids also have a number of carboxylate groups that alsocoordinate to the Ln3+ ion. Early experiments on separating lanthanides were carried outon a tracer scale using citrate complexes, but EDTA and related ligands were found togive lanthanide complexes that afforded better separations on a large scale (Section 1.5);thus log K for the EDTA complexes of Pr3+and Nd3+are 16.40 and 16.61, respectively;the average difference in K between adjacent lanthanides is 100.4. EDTA is a hexadentateligand; since transition metals can also coordinate a water molecule to form complexes like[Fe(EDTA)(H2O)]−, it is not surprising that lanthanides form complexes with more watermolecules bound. The stoichiometry of the complex isolated depends to some extent uponthe counter-ion; for example, nine-coordinate [Ho(EDTA)(H2O)3]− ions are found in thecrystal when M = K, whilst eight-coordinate [Ho(EDTA)(H2O)2]− ions are found whenM = Cs. In general, the early lanthanides form M[Ln(EDTA)(H2O)3] and the later ones[M(EDTA)(H2O)2]. {See Figure 4.5 for the [La(EDTA)(H2O)3]− complex.}

Diethylenetriaminepentaacetic acid (H5DTPA) is a potentially octadentate ligand (re-alized in practice) and forms complexes with larger stability constants, of the type[M(DTPA)(H2O)]2−. DTPA complexes have greater stability constants than do EDTA

...


Complexes 43

Figure 4.5The structure of La(EDTA)(H2O) −

3 (reproduced by permission of Macmillan from S.A. Cotton,Lanthanides and Actinides, 1991, p. 64).

complexes, since the greater denticity of the ligand makes it harder for the metal ion tobe removed. Another factor is that the entropy change on complex formation is likely to begreater, thus making �G (and hence K) greater. Compare

[La(OH2)9]3+(aq) + EDTA4−(aq) → [La(EDTA)(OH2)3]−(aq) + 6 H2O(l)

[La(OH2)9]3+(aq) + DTPA5−(aq) → [La(DTPA)(OH2)]2−(aq) + 8 H2O(l)

The synthesis and study of gadolinium complexes as potential magnetic resonance imag-ing agents is of intense interest at the moment, and is discussed in the section on MRI inSection 5.5.2.

4.3.9 Complexes of N-Donors

Although pyridine and similar bases are well known to form adducts with β-diketonatecomplexes, relatively few complexes of monodentate N-donors exist. Choice of solvent torestrict the oxophilic tendencies of the lanthanides is important. Thus, ammine complexes,[Yb(NH3)8][Cu(S4)2].NH3, [Yb(NH3)8][Ag(S4)2].2NH3, and [La(NH3)9][Cu(S4)2] havebeen synthesized by reactions in supercritical ammonia. Direct reaction of the lanthanidehalides with pyridine gives pyridine complexes of the lanthanides, with the synthesis of[YCl3py4] and [LnCl3py4].0.5py (Ln = La, Er). These all have pentagonal bipyramidalstructures, with two chlorines occupying the axial positions. [MI3py4] (M = Ce, Nd) andEuCl3py4 have also been reported, whilst square antiprismatic [Sm(meim)8]I3 and a fewother N-methylimidazole complexes such as [YX2(N-meim)5]+ X− (X = Cl, Br) have alsobeen made.

In the case of thiocyanate complexes, a wide variety of species of known struc-ture have been made, including (Bu4N)3[M(NCS)6] (M = Y, Pr–Yb; octahedral co-ordination); (Et4N)4 [M(NCS)7].benzene (M = La, Pr: capped trigonal prismatic);(Me4N)4 [M(NCS)7].benzene (M = Dy, Et, Tb: pentagonal bipyramidal); and (Me4N)5

[M(NCS)8].benzene (M = La–Dy: intermediate between square antiprism and cubic),showing the dependence of coordination number in the solid state upon factors such asthe counter-ion.

The best examples of complexes with bidentate N-donors are those with 2,2′-bipyridyl(bipy) and 1,10-phenanthroline (phen). Complexes LnL2(NO3)3 (L = bipy, phen) havebeen studied in detail and have 10-coordinate structures with all nitrates bidentate

...



Figure 4.6The structure of La(phen)2(NO3)3 (reproduced by permission of the American Chemical Society fromM. Frechette, I.R. Butler, R. Hynes, and C. Detellier, Inorg. Chem., 1992, 31, 1650).

(Figure 4.6). Ln(bipy)3(NO3)3 are in fact Ln(bipy)2(NO3)3.bipy (X-ray for Nd). Thiocy-anates LnL3(NCS)3 (Ln = Pr; L = phen, bipy) have nine-coordinate monomeric structures,whilst chloride complexes with a range of structures include [Nd(bipy)2Cl3(OH2)].EtOHand [Yb(bipy)2Cl3].

Tridentate N-donor ligands are efficient in separating actinides from lanthanides selec-tively by solvent extraction, an area of potential great importance in treatment of usednuclear fuel rods. The tridentate ligand 2,2’: 6’,2” -terpyridyl (terpy) forms a range of com-plexes. The perchlorate complexes [Ln(terpy)3] (ClO4)3 contain nine-coordinate cationswith near-D3 symmetry, a structure initially deduced from the fluorescence spectrum of theeuropium compound (Section 5.4) and subsequently confirmed by X-ray diffraction studies(Figure 4.7)

A number of chloride complexes include Pr(terpy)Cl3.8H2O; X-ray diffraction estab-lished the presence of a [Pr(terpy)Cl(H2O)5]2+ ion. Lanthanide nitrates react with 1 moleof terpy in MeCN to give [La(terpy)(NO3)3(H2O)n] (Ln = La, n = 2; Ln = Ce–Ho, n = 1;Ln = Er–Lu, n = 0) with coordination numbers decreasing from 11 (La) to 9 (Er–Lu);if these are crystallized from water [Ln(terpy)(NO3)2(H2O)4] NO3 (Ln = La–Gd) and[Ln(terpy)(NO3)2(H2O)3]NO3.2H2O (Ln = Tb–Lu) are formed; these have both biden-tate and ionic nitrate and nine- and eight-coordinate lanthanides, respectively. Thiocyanatecomplexes Ln(terpy)2(NCS)3(Ln = Pr, Nd) are also nine coordinate.

4.3.10 Complexes of Porphyrins and Related Systems

One convenient synthesis of these compounds involves heating La(acac)3 withH2TPP (H2TPP = tetraphenylporphyrin) and other porphyrins in high-boiling solventssuch as 1,3,5-trichlorobenzene (TCB). Products depend upon conditions and include[Ln(acac)(tpp)] and the double-decker sandwich [Ln2(tpp)3]. Less forcing routes can beused, for example the reaction of [Y{CH(SiMe3)2}3] and yttrium alkoxides with H2OEP

...


Complexes 45

Figure 4.7The structure of Eu(terpy) 3+

3 (reproduced by permission of the Royal Society of Chemistry from G.H.Frost, F.A. Hart, C. Heath, and M.B. Hursthouse, Chem. Commun., 1969, 1421).

(octaethylporphyrin) to form compounds like [Y{CH(SiMe3)2}(Oep)]. Bis(porphyrin) com-plexes and mixed (porphyrin)phthalocyanine complexes are probably complexes of thetrivalent lanthanide where one ligand is a dianion and another a one-electron-oxidizedπ -radical, as in [LnIII(pc)(tpp)] (Ln = La, Pr, Nd, Eu, Gd, Er, Lu, Y).

Complexes, particularly of gadolinium, with a N4-donor macrocycle backbone that alsosupports carboxylate groups have assumed importance in the context of MRI contrast agents(Sections 5.5.2 and 5.5.3).

Gadolinium texaphyrin (Gd-tex; Figure 4.8) is reported to be an effective radiation sen-sitizer for tumour cells, whilst the corresponding lutetium compound, which absorbs lightin the far-red end of the visible spectrum, is in Phase II trials for photodynamic therapy forbrain tumours and breast cancer.

O O OMe

O OMe

O

OO

OH

Gd

OH

N

N

NN

N

OAcAcO

Figure 4.8Gd-tex.

...



4.3.11 Halide Complexes

Complexes with all the halide ions have been made, though the fluorides and chloridesare the most studied. Fluorides ALnF4, A2LnF5 and A3LnF6 exist (A = alkali metal); mosthave structures with 8- or 9-coordinate metals, though a few A3LnF6 have the six-coordinatecryolite structure.

Cs2LiLnCl6 are an important type of complex.

2 CsCl + LiCl + LnCl3 → Cs2LiLaCl6

Made by fusion, they possess the six-coordinate elpasolite structure in which the perfectlyoctahedral [LnCl6]3− ions occupy sites with cubic symmetry. These have been much studiedfor their optical and magnetic properties on account of this, the Ln3+ ions on their highlysymmetrical sites having especially weak f–f transitions in their electronic spectra. Someother A2BMX6 systems also adopt the elpasolite structure. Other halide complexes like(Ph3PH)3(LnX6) and (pyH)3(LnX6) (X = Cl, Br, I) have been made by solution methods:

LnX3 + 3Ph3PHX → (Ph3PH)3LnX6 (X = Cl, Br; in EtOH)

(Ph3PH)3SmBr6(s) + 6 HI(l) → (Ph3PH)3SmI6(s) + 6 HBr(l)

The iodides are hygroscopic and rather unstable. Other stoichiometries are possible:

LiCl + GdCl3 → LiGdCl4; 4 KI + 2 Pr + 3 I2 → 2 K2PrI5

including species with more than one lanthanide ion:

3 CsI + 2 YI3 → Cs3Y2I9

Six coordination is the norm for chlorides, bromides, and iodides, but it is not invariable.Formulae are not necessarily a guide to the species present; thus, octahedral [YbCl6]3− ionsare present in M4YbCl7 (M = Cs, MeNH3) but Ba2EuCl7 contains capped trigonal prismatic[EuCl7]4− ions. Compounds like Cs2Y2I9 attain six coordination through face-sharing ofoctahedra, Cs2DyCl5 has corner-sharing of octahedra, and M2PrBr5 (M = K, NH4) obtainseven coordination though edge-sharing. Structure type can depend on temperature; thus,NaLnCl4 (Ln = Eu–Yb, Y) have the six-coordinate α-NiWO4 structure at low temperatureand the seven-coordinate NaGdCl4 structure at high temperatures. To summarize, althoughrewarding study over many years for the variety of structures adopted and the opportunitiesto study optical and magnetic properties, the area of complex halides is complicated andeach case has to be judged on its own merits.

4.3.12 Complexes of S-Donors

As a ‘soft’ donor, sulfur is not expected to complex well with lanthanides; nevertheless anumber of complexes have been isolated through working in nonpolar solvents. The simplestof these complexes are the dithiocarbamates Ln(S2CNR2)3 and related dithiophosphatesLn(S2PR2)3; of the latter, Ln[S2P(cyclohexyl)2]3(Ln = Pr, Sm) have trigonal prismatic ge-ometries. The dithiocarbamates form more stable 8-coordinate adducts Ln(S2CNR2)3.bipy(R = Me, Et) and anionic species [Ln(S2CNR2)4]−. A range of geometries are known in thesolid state: distorted dodecahedral Et4N[Eu(S2CNEt2)4], Na[Eu(S2CNEt2)4], near-perfectdodecahedral (Ph4As)[Ln(S2PEt2)4], and square antiprismatic Ph4P [Pr(S2PMe2)4].

A number of lanthanide thiolates have been synthesized in recent years and are discussedin Section 4.4.3.

...


Alkoxides, Alkylamides and Related Substances 47

4.4 Alkoxides, Alkylamides and Related Substances

Some of these compounds have been known for over 30 years, but interest in the former inparticular has been stimulated recently by their potential use as precursors for depositionof metal oxides using the sol-gel or MOCVD process. These compounds are particularlyinteresting insofar as many of them involve very bulky ligands and therefore have lowcoordination numbers (see Section 4.5).

4.4.1 Alkylamides

The volatile air-sensitive silylamides Ln[N(SiMe3)2]3 are the best-characterized alky-lamides. They are made by salt-elimination reactions in a solvent like THF:

LnCl3 + 3 LiN(SiMe3)2 → 3 LiCl + Ln[N(SiMe3)2]3

They are volatile in vacuo at 100 ◦C and dissolve in hydrocarbons; the first three-coordinatelanthanide compounds to be characterized, they have congested three-coordinate geometries(Figure 4.9) due to the bulky SiMe3 groups. They are pyramidal in the solid state with N–Ln–N angles around 114◦ rather than the 120◦ expected for a planar structure but they areevidently planar in solution as they have no dipole moment. Theoretical calculations (D.L.Clark et al., 2002; L. Penin et al., 2002) indicate that it is β-Si–C agostic interactions withthe lanthanide that are responsible for this small pyramidal distortion.

Ln[N(SiMe3)2]3 form stable four-coordinate Ph3PO and Ph2CO adducts; when ex-cess of phosphine oxide is used, five-coordinate peroxide-bridged dimers are obtained.Eu[N(SiMe3)2]3 forms both 1:1 and 1:2 adducts with Me3PO. They are capable of bindingtwo very small MeCN groups, the latter reversibly, but no thf adducts have been isolated.

When a deficit of the lithium alkylamide is used, dimeric chloride-bridged silylamidesare obtained which are useful synthetic intermediates (Figure 4.10).

Using less bulky alkyl substituents, compounds [Ln(N(SiHMe2)2)3(thf)2] have been madefor La–Lu, Y. These have trigonal bipyramidal structures with axial thf molecules {the scan-dium analogue is pseudo-tetrahedral Sc[N(SiHMe2)2]3(thf)}. Other alkylamides Ln(NPri

2)3

(Ln = La, Nd, Yb, Y, Lu) are obtained with the less hindered isopropylamide ligand; thesecompounds form isolable thf adducts. Simple dimethylamides cannot be obtained, as reac-tion of LnCl3 with LiNMe2 gives Ln(NMe2)3 as LiCl adducts.

Use of a highly fluorinated amide has permitted the synthesis of [Sm[N(SiMe3)(C6F5)]3]in which there are many Sm. . . .F and agostic interactions (Figure 4.11), so that the truecoordination number of the metal is really greater than three.

Me

N

SiMe3

SiMe3

SiMe2

SiMe2

Me2Si

Me3Si

N

MeMe

Sm

N

Figure 4.9The structure of Ln[N(SiMe3)2]3.

...



2

(Ln = Lu)

Figure 4.10Lanthanide bis(trimethylsilyl)amido complexes.

N

F

F

F

F

F

Si

MeMe Me

Sm

Me

MeMe

Si

F

F

F

F

F

N

Me

Me

MeSi

F

F F

F

FN

Figure 4.11The structure of [Ln[N(SiMe3)(C6F5)]3].

The neodymium compound [(η6-C6H5Me)Nd{N(C6F5)2}3] has a η6-bonded toluenemolecule with a distorted piano-stool geometry.

4.4.2 Alkoxides

The simplest lanthanide alkoxide compounds about which much is known are the iso-propoxides. There is some evidence that under mild conditions a simple Ln(OPri)3 can be

...


Alkoxides, Alkylamides and Related Substances 49

Figure 4.12The structure of Y5O(OPri)13 (reproduced by permission of the American Chemical Society fromO. Poncelet, O.W.J. Sartain, L.G. Hubert-Pfalzgraf, K. Folting, and K.G. Caulton, Inorg. Chem.,1989, 28, 264).

formed, but the product generally obtained from synthesis [which may be described by thesynthesizers as ‘Ln(OPri)3’] is an oxo-centred cluster Ln5O(OPri)13 (Figure 4.12) (X-rayfor Y, Er, Yb, etc.) with a square-pyramidal core.

Syntheses include:

NdCl3(PriOH)3 + 3 KOPri → Nd(OPri)3 + 3 KCl + 3 PriOH

Y + 3 PriOH → Y(OPri)3 + 3/2 H2

Pr[(N(SiMe3)2]3 + 3 PriOH → Pr(OPri)3 + 3 HN(SiMe3)2

The reaction conditions for syntheses seem important, as a chlorine-containing clusterNd6(OPri)17Cl has also been reported, from the reaction of NdCl3 with NaOPri. Alkoxidesof known structure include Ln(OCH2But )3, which is a tetramer (Ln = La, Nd), based ona square of lanthanide atoms with bridging alkoxides and no direct metal–metal bonds[Ln4(µ2-OR)8(OR)4] (R = CH2But ). La(OBut )3 was isolated as a trimer containing atriangle of lanthanum atoms (and two coordinated alcohol molecules) [La3(µ3-OBut )3(µ2-OBut )3(OBut )4(But OH)2]; Nd(OCHPri)3 has been isolated as a dimeric solvate [Nd2(µ2-OR)2(OR)4L2] (L = py, thf). Ce(OBut )3 is believed to be a monomer. It undergoes ther-molysis in high yield.

Dimeric structures are also exhibited by [Ln(OQPh)3]2 (Ln = La, Ce, Y; Q = C, Si),though the bridges are cleaved by Lewis bases, forming adducts such as Ln(OSiPh)3(thf)3

(fac-octahedral; Ln = Y, La, Ce) and Y(OSiPh)3(OPBu3)2 (tbp).A range of aryloxides has been made. The diphenylphenoxides Ln(OR)3 (R = 2,6-

Ph2C6H3O) have three-coordinate pyramidal structures with some additional Ln–ring in-teractions. When the bulk of the aryloxide is increased by putting two tert-butyl groups intothe 2,6-positions of the benzene ring, three-coordinate aryloxides Ln(O-2,6But

2-4-RC6H2)3

can be isolated (R = H, Ln = Y, La, Sm; R = Me, Ln = Y, La, Ce, Pr, Nd, Dy–Er, Yb). Trig-onal planar geometry is found in Y(O-2,6-But

2C6H3)3 but a pyramidal structure is adoptedin Ce(O-2,6-But

2-4-MeC6H2)3. Synthesis of Nd(Odpp)3 in THF gives [Nd(Odpp)3(thf)2],which crystallizes from toluene as [Nd(Odpp)3(thf)] and sublimes as [Nd(Odpp)3]. Thestructure of [Nd(Odpp)3(thf)2].2THF shows it to be a five-coordinate bis(thf) complex inwhich the five ligands bind only by oxygen. Removal of the thf ligands affords compounds

...



where benzene rings additionally bond to the lanthanide. The departure of one or twoelectron-pair donors from the coordination sphere means that the Ln3+ ion accepts electrondensity from ligand π systems and in a sense ‘maintains its coordination number’. (see G.B.Deacon et al., J. Chem. Soc., Dalton Trans., 2000, 961 for a fuller account of these unusualcompounds.)

There is a more limited chemistry of alkoxides and aryloxides in the +2 and +4 states.The CeIV compounds, the only ones known in this oxidation state, are discussed in Section4.6.2.

Much of the chemistry in the +2 state has been with aryloxides; thus 2,6-diphenylphenol(dppOH) reacts with Eu or Yb on heating in the presence of mercury, forming [Ln(Odpp)2].These have the structures [Eu2(Odpp)(µ-Odpp)3] and [Yb2(Odpp)(µ-Odpp)3].

Another route involves alcoholysis of an aryl (R = 2,6-But2C6H3O):

(C6F5)2YbROH−→THF

Yb(OR)2(thf)n (n = 2, 3)

(C6F5)2YbROH−→Et2O

Yb(OR)2(Et2O)2

In recent years, phosphide complexes have proved accessible:

YbI2(thf)2 + 2 KPPh2 → trans-Yb(PPh2)2(thf)4 + 2 KI

Yb(PPh2)2(thf)4 + 4 MeIm → trans-Yb(PPh2)2(meim)4 + 4 THF

(meim = 1 − methylimidazole)

Similar compounds have been synthesized, including trans-Sm(Qmes2)2(thf)4(Q = P,As; mes = mesityl); [Nd{P(SiMe3)2}3(thf)2] has a trigonal bipyramidal structure.

4.4.3 Thiolates

Lanthanide thiolates are rare, but monomeric species have been obtained using bulkyligands, such as the three-coordinate Sm(S-2,4,6-But

3C6H2)3 and the monomeric, mer-[Yb(SPh)3py3]. Ln(SPh)3 reacts with S, forming octanuclear clusters [Ln8S6(SPh)12(thf)8](Ln = Ce, Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er); analogous pyridine clusters [Ln8S6(SPh)12(py)8](Ln = Nd, Sm, Er) have also been isolated. Ph2Se2 reacts with lanthanide amalgams andpyridine, forming Ln(SePh)3(py)3 (Ln = Ho, Tm, Yb) which are dimeric with 7-coordinatelanthanides. [Ln(SePh)3(thf)3] (Ln = Tm, Ho, Er) have monomeric fac-octahedral struc-tures; [Ln8E6(EPh)12L8] clusters (E = S, Se; Ln = lanthanide; L = Lewis base) can beprepared by reduction of Se–C bonds by low-valent Ln or by reaction of Ln(SePh)3. Struc-tures reported include [Sm8E6(SPh)12(thf)8] (E = S, Se) and [Sm8Se6(SePh)12(py)8]; theyhave cubes of lanthanide ions with E2− ions capping the faces and EPh bridging the edgesof the cube. Reaction of Nd(SePh)3 with Se to form [Nd8Se6(SePh)12(py)8] shows that thisseries is not restricted to redox-active lanthanides.

4.5 Coordination Numbers in Lanthanide Complexes

4.5.1 General Principles

Ionic radii of La3+ and Lu3+ ions in octahedral coordination are 1.032 A, and 0.861 A,respectively. The corresponding value for the largest M3+ ion of a transition metal is 0.670 A

...


Coordination Numbers in Lanthanide Complexes 51

for Ti3+. Purely on steric grounds, therefore, it would be expected that lanthanide ions couldaccommodate more than six ligands in their coordination sphere. Coupled with this is thefact that the f-orbitals are ‘inner’ orbitals, shielded from the effects of the surroundinganions and therefore not able to participate in directional bonding; there are none of theligand-field effects found in transition metal chemistry with the concomitant preference foroctahedral coordination. In summary, the coordination number (CN) adopted by a particularcomplex is determined by how many ligands can be packed round the central metal ion;coordination numbers between 2 and 12 are known in lanthanide complexes. Geometriesare those corresponding to the simple polyhedra that would be predicted from electron-pair repulsion models; tetrahedral (CN 4), trigonal bipyramidal (CN 5), octahedral (CN 6);though with coordination numbers 2 and 3 deviations from simple linear and trigonal planargeometries in the solid state often occur (and are probably due to agostic interactions).

Some 20 years ago, analysis of the coordination numbers for large numbers of coordina-tion compounds of yttrium and the lanthanides indicated that coordination numbers of 8 and9 are almost equally common, accounting for around 60% of the known structures, and it isunlikely that this distribution has changed significantly. As already pointed out, it is stericfactors that determine the coordination number (and geometry) adopted by a lanthanideion (see e.g. X.-Z. Feng, A.-L. Guo, Y.-T. Xu, X.-F. Li and P.-N. Sun, Polyhedron, 1987,6, 1041; J. Marcalo and A. Pires de Matos, Polyhedron, 1989, 8, 2431). Saturation in thecoordination sphere of the metal can come about in one of two ways.

First-Order Effects

When small ligands like water or chloride bind to a metal, the coordination number isdetermined by how many ligands can pack round the central metal ion, a number relating torepulsion between the donor atoms directly in contact with the metal, a so-called ‘first-order’effect.

Second-Order Effects

Certain ‘bulky’ ligands have a small donor atom (N, O, C) attached to bulky sub-stituents, examples being certain alkoxides and aryloxides (and related systems),bis(trimethylsilyl)amido [-N(SiMe3)2], and the isolobal alkyl [-CH(SiMe3)2]; and mesityl.These ligands have high second-order steric effects generating crowding round the lan-thanide; even though the metal ion is bound to few donor atoms, the bulk of the rest of theligand shields the metal from other would-be ligands.

Species like LnCl3−6 and [Ln(H2O)9]3+ are examples where first-order effects deter-

mine the coordination number, whilst Ln[CH(SiMe3)2]3 and [Ln{N(SiMe3)2}3.(Ph3PO)]are cases where the bulky substituents on the amide and alkyl ligands with high second-ordersteric effects generate crowding round the lanthanide.

4.5.2 Examples of the Coordination Numbers

Coordination Number 2

Coordination number 2 is represented by the sublimeable two coordinate bent alkyl[Yb{C(SiMe3)3}2] (C–Yb–C 137◦) and the europium analogue. Although there are onlytwo Yb–C σ bonds, the bending is caused by a number of agostic Yb. . . H–C interactions

...



(a similar situation applying to some zerovalent Pd and Pt phosphine complexes M(PR3)2,where R is a bulky group like But or cyclohexyl).


This is best illustrated by some alkyl and alkylamide LnX3 systems {X = a bulky groupsuch as -N(SiMe3)2 ,-CH(SiMe3)2, e.g. Ln[CH(SiMe3)2]3 (Ln = Y, La, Pr, Nd, Sm, Lu);Ln[N(SiMe3)2]3, (Ln = Y, all lanthanides)}. Many transition metals form analogous, planar,three-coordinate alkylamides; the majority of these lanthanide compounds have pyramidalstructures in the solid state [but are generally planar in the gas phase and in solution (theyhave zero dipole moment)]. These distortions appear to involve agostic interactions. Certainalkoxides and aryloxides Ln(OR)3 are formed using bulky ligands (e.g. Ln = Ce; R = 2,6-But

2C6H3O).


A few four-coordinate compounds are known, such as Ln[(N(SiMe3)2]3.(Ph3PO), [Li(thf)4][Ln(NPh2)4] (Ln = Er, Yb), [Li(thf)4] [Lu(2,6-dimethylphenyl)4], and Li(THF)4][Ln(CH2

SiMe3)4] (Ln = Y, Er, Tb, Yb); there is a tetrahedral geometry as expected. Again, the ligandshave high second-order steric effects generating crowding round the lanthanide even thoughthere are only four atoms bound to the lanthanide.


As yet this is a rare coordination number. The three-coordinate silylamides can add twoslender ligands to form five-coordinate examples Ln[N(SiMe3)2]3(NCMe)2, whilst thereare a few alkyls e.g. Yb(CH2But )3(thf)2.[Nd{P(SiMe3)2}3(thf)2] has also been described.These have trigonal bipyramidal structures, as expected.


Salts Cs2LiLnCl6 (Ln = La–Lu) have the elpasolite structure with [LnCl6]3− anions on cubicsites; this geometry could be due to first-order crowding among the six chlorides, but someother salts are known where LnCl7 units occur. Other examples include the reactive alkyls[Li(L-L)3] [Ln(CH3)6] (L-L = MeOCH2CH2OMe or Me2NCH2CH2NMe2; Ln = La–Sm,Gd–Lu, Y) and diketonates of bulky ligands such as [Ln(But COCHCOBut )3] (Ln = Tb–Lu).Some compounds have geometry distorted to trigonal prismatic because of steric interactionsbetween ligands, such as [Ln{S2P(cyclohexyl)2}3]. Some thiocyanate complexes (Bu4N)3

[M(NCS)6] (M = Y, Pr–Yb) have octahedral coordination (Section 4.3.9) but others like(Et4N)4 [M(NCS)7].benzene (M = La, Pr) and (Me4N)5 [M(NCS)8].benzene (M = La–Dy) are 7- and 8-coordinate, respectively, a reminder that the coordination number adopteddepends upon a subtle balance between factors, including the counter-ion and solvent.


The most common geometries encountered are capped octahedral and capped trigonalprismatic. Many of the known seven-coordinate compounds involve β-diketonate ligands,particularly adducts of the type Ln(diketonate)3.L (L = Lewis base e.g. H2O, py). Several

...


Coordination Numbers in Lanthanide Complexes 53

thf complexes (Section 4.3.3) of the type Ln(thf)4X3 (X = Cl, NCS) adopt pentagonalbipyramidal geometries), as does the ytterbium(II) aryl [Yb(C6F5)2(thf)5]. Ba2EuCl7 hascapped trigonal prismatic [EuCl7]4− ions. The energy difference between these two geome-tries is small, witness (Et4N)4 [M(NCS)7].benzene (M = La, Pr: capped trigonal prismatic)and (Me4N)4 [M(NCS)7].benzene (M = Dy, Et, Tb: pentagonal bipyramidal).


Most eight-coordinate lanthanide complexes have dodecahedral or square antiprismatic ge-ometries. The energy difference between these is likely to be small, witness the distorteddodecahedral Et4N [Eu(S2CNEt2)4], near-perfect dodecahedral (Ph4As)[Ln(S2PEt2)4],and square antiprismatic Ph4P [Pr(S2PMe2)4]. Similarly, among eight-coordinate thio-cyanate complexes, (Me4N)5 [M(NCS)8].benzene (M = La–Dy) are intermediate be-tween square antiprism and cubic; (Et4N) [M(NCS)4(H2O)4] (M = Nd, Eu) and (Me4N)3

[M(NCS)6(H2O)(MeOH)] (M = La–Dy, Er) are square antiprismatic; (Et4N)4[M(NCS)7

(H2O)] (M = La–Nd, Dy, Er) are cubic; and (Me4N)5 [M(NCS)8].benzene (M = La–Dy)are intermediate between square antiprismatic and cubic.


Tricapped trigonal prismatic is the most familiar example of nine-coordinate geometry,adopted for the [Ln(H2O)9]3+ ions of all lanthanides in a number of crystalline salts. Itis also found in a number of chlorides LnCl3 (Ln = La–Gd) and bromides LnBr3 (Ln =La–Pr). This geometry is also sometimes adopted where polydentate ligands are involved,such as the [Ln(terpy)3]3+ ion (terpy = 2,2’:6’,2”-terpyridyl; see Figure 4.7).

Coordination Numbers 10–12

Sheer congestion of donor atoms around the metal ion and concomitant inter-donor atomrepulsions makes these high coordination numbers difficult to attain. They are often as-sociated with multidentate ligands with a small ‘bite angle’ such as nitrate that take uplittle space in the coordination sphere, either alone, as in (Ph4As)2[Eu(NO3)5] or in combi-nation with other ligands, as in Ln(bipy)2(NO3)3, Ln(terpy)(NO3)3(H2O) (Ln = Ce–Ho),and crown ether complexes (Section 4.3.7) such as Ln(12-crown-4)(NO3)3(Ln = Nd–Lu).Other crown ether complexes can have 11 and 12 coordination, e.g. Eu(15-crown-5)(NO3)3

(Ln = Nd–Lu) and Ln(18-crown-6)(NO3)3(Ln = La, Nd).Polyhedra in these high coordination numbers are often necessarily irregular, but when all

the ligands are identical, near-icosahedral geometries occur for the 12-coordinate [Pr(1,8-naphthyridine)6]3+ and [La(NO3)6]3− ions in crystalline salts. It should also be rememberedthat the geometries discussed here are found in the solid state, but on dissolution in a solvent,where the influence of counter-ions is lessened, matters may be different (see the aqua ions,Sections 4.3.1 and 4.3.2). In principle, isomers are often possible, but because of the labilityof lanthanide complexes they are very rarely observed.

4.5.3 The Lanthanide Contraction and Coordination Numbers

Certain generalizations can be made. As noted elsewhere (Section 2.4) the contraction ofthe (5s2 5p6) configuration with increasing nuclear charge, due to penetration inside the

...



f-orbitals, means that the Ln3+ (and Ln2+) ions become smaller as the atomic numberincreases. Because of the decrease in ionic radius as the atomic number of the lanthanideincreases, it would be expected that fewer anions could be packed round the central metalion as the ionic radius decreases; in other words, that the coordination number will decreasewith increasing Z. It may similarly be predicted that the coordination number would decreasewith increasing size of the anion. This behaviour is clearly observed in the halides and oxides(Chapter 3). To reiterate, in the fluorides LnF3, the coordination number decreases from11 in LaF3 through 9 in SmF3; similarly in the trichlorides, lanthanum is 9 coordinate inLaCl3 but lutetium is 6 coordinate in LuCl3. The coordination number of lutetium decreasesfrom 9 in LuF3 to 6 in LuI3. Similar changes in coordination number are found in manycomplexes, the best known example being the hydrated Ln3+(aq) ion. Nine-coordinate[Ln(OH2)9]3+ ions are found in solution for the earlier metals (Ln = La–Sm/Eu) and eight-coordinate [Ln(OH2)8]3+ ions for later metals (Dy–Lu, Y) with a mixture of species for theintermediate metals.

This decrease in coordination number is not an invariable rule, as examples are knownwhere the coordination number remains constant across the series, as in Ln(bipy)2(NO3)3

and Ln(phen)2(NO3)3 (Ln = La–Lu) where all are 10 coordinate in the solid state, orLn(hmpa)3Cl3, all of which appear to be 6 coordinate. When bidentate ligands are involved,loss of one of them may produce a too dramatic change in CN. The influence of a counter-ion, which will affect the lattice energy and hence the solubility, determining which com-plex species crystallizes first, can also be important. The best example of this lies in thecrystalline hydrated salts of several oxoacids [Ln(OH2)9]X3 (Ln = La–Lu, Y; X e.g. BrO3,CF3SO3, C2H5SO4) all of which contain the [Ln(OH2)9]3+ ion, whereas the coordinationnumber of the Ln3+(aq) ion varies in solution, as noted above. In contrast, the perchloratesLn(ClO4)3.6H2O contain octahedral [Ln(OH2)6]3+ ions. Similarly, in the edta complexesM[Ln(edta)(H2O)x ], nine-coordinate [Ho(edta)(H2O)3]− ions are found in the crystal whenM = K whilst eight-coordinate [Ho(edta)(H2O)2]− ions are found when M = Cs.

4.5.4 Formulae and Coordination Numbers

Many transition metals adopt characteristic coordination numbers in their compounds,and consequently it is often possible to deduce whether (and how) a potential ligandwill coordinate or not. The lanthanides do not have characteristic coordination num-bers, so such assumptions are not possible; ultimately, X-ray diffraction studies arethe only sure guide. The presence of coordinated solvent molecules in crystals ob-tained from polar solvents is generally best deduced thus, hence the assignment of an8-coordinate structure to [LaCl3(phen)2(H2O)].MeOH. Likewise La(phen)3Cl3.9H2O hasbeen shown to be [La(phen)2(OH2)3]Cl3.4H2O.phen. [Sm(NO3)3(Ph3PO)3].2 acetone and[Sm(NO3)3(Ph3QO)2(EtOH)].acetone (Q = P, As) have all been shown to have nine-coordinate samarium.

4.6 The Coordination Chemistry of the +2 and +4 States

These areas have remained relatively undeveloped until recently; with increasing emphasison the use of nonaqueous solvents, many discoveries in the (+2) state for Eu, Sm, and Ybin particular probably remain to be made.

...


The Coordination Chemistry of the +2 and +4 States 55

4.6.1 The (+2) State

Solutions of Eu3+(aq) are reduced by zinc to Eu2+(aq); this can be precipitated as thesulfate, affording a separation from the other Ln3+(aq) ions, which were not reducedunder these conditions, used for many years in the purification of the lanthanides. Ingeneral the solubilities of the ‘inorganic’ compounds of the Ln2+ ions resemble thoseof the corresponding compounds of the alkaline earth metals (insoluble sulfate, carbon-ate, hydroxide, oxalate). Electrolytic reduction is one of a number of methods that canbe used to generate solutions of Eu2+, Sm2+, and Yb2+ ions; aqueous solutions of Sm2+

and Yb2+ tend to be short-lived and oxygen-sensitive in particular. These and other di-valent lanthanides can be stabilized by the use of nonaqueous solvents such as HMPAand THF, in which they have characteristic colours, different and deeper than those in theisoelectronic Ln3+ ions on account of the decreased term separations in the divalent ions(typically, Eu2+ pale yellow; Sm2+ red; Yb2+ green-yellow; Tm2+ green; Dy2+ brown;Nd2+ red).

A significant number of complexes are now known, including SmCl2(thf)5 andEuI2(thf)5 (pentagonal bipyramidal); polymeric [SmCl2(But CN)2]∞ (six coordinate);LnI2(N-methylimidazole)4 (Ln = Sm, Eu); EuCl2(phen)2 and EuCl2(terpy); trans-EuI2(thf)4; [Yb(hmpa)4(thf)2]I2, [Sm(hmpa)6]I2,[Sm(hmpa)4I2] and the eight-coordinate[SmI2{O(C2H4OMe)2}] (the latter existing as cis and trans isomers). Compounds of lan-thanides with an accessible Ln2+ ion yet previously unknown coordination chemistry arestarting to be made, such as [LnI2(dme)3] and [LnI2(thf)5] (Ln = Nd, Dy), made by re-action of Nd and Dy with I2 at 1500 ◦C, followed by dissolution of the product in theappropriate ligand. [TmI2(dme)3] has seven-coordinate thulium, with one monodentatedimethoxyethane; the samarium analogue, which contains a larger metal ion, has eightcoordination.

The bis(silylamide) complexes Ln[N(SiMe3)2]2 (Ln = Eu, Yb) are well characterized;because of the bulky ligands, these tend to exhibit three and four coordination, much asfor the corresponding compounds in the +3 oxidation state. Thus Yb[N(SiMe3)2]2 is adimer with two bridging alkylamides, and the [Eu{N(SiMe3)2}2]− ion is trigonal planar(unlike the EuIII analogue). These compounds form adducts with monomeric structures;M[N(SiMe3)2]2L2 (M = Eu, Yb; L, e.g., PBu3, THF; L2 = dmpe) are four-coordinate, andEu[N(SiMe3)2]2(glyme)2 is six-coordinate.

There is a developing chemistry with alkoxide and aryloxide as well as amide lig-ands, together with organometallics. To give some examples from ytterbium(II) aryloxides,Yb(Odpp)2(thf)3 (Odpp = 2,6-diphenylphenoxide) is trigonal bipyramidal, whilst with thevery bulky 2,6-di-tert-butyl-4-methylphenoxy (Odtb) ligand the adduct Yb(Odtb)2(thf)2

is tetrahedral whilst Yb(Odtb)2(thf)3 is five-coordinate square pyramidal. Among com-pounds with the heavier Group VI (16) donor atoms, Yb(SAr)2(MeOCH2CH2OMe)2andLn(Qmes)2(thf)n] (Ln = Yb, Sm, Eu; Q = Se, Te; mes = 2,4,6-trimethylphenyl) are ex-amples. The most striking organometallics are the two-coordinate alkyls [Ln[C(SiMe3)3]2](Ln = Yb, Eu), the aryls LnPh2(thf)n(Ln = Eu, Yb), with as-yet unknown structures,Yb(C6F5)2(thf)4 of unknown structure, and Eu(C6F5)2(thf)5 having a pentagonal bipyra-midal structure, with all thf equatorial. The synthesis has also been reported of tetrahedral[Eu(dpp)2(thf)2] (dpp = 2,6-diphenylphenyl). Much of this work is very recent, thoughcyclopentadienyl-type systems have long been known in the (+2) state. Organometallicsare discussed in more detail in Chapter 6.

...



4.6.2 The (+4) State

Cerium is the only lanthanide with an extensive chemistry in this oxidation state. Thecerium (+4) aqua ion, obtained by acid dissolution of CeO2, is thermodynamically unsta-ble but kinetically stable; the anion sensitivity of the reduction potentials (E = 1.61 V in1M HNO3; 1.44 V in 1M H2SO4; and 1.70 V in 1M HClO4) indicates that complexes areinvolved rather than just a simple aqua ion. Salts are known, including CeSO4.4H2O andCe(NO3)4.5H2O; the latter probably contains 11-coordinate [Ce(NO3)4.(H2O)3] molecules,like the thorium analogue. Cerium is 10 coordinate in the [Ce(CO3)5]6− ion and 12 coor-dinate in [Ce(NO3)6]2−, the ion found in the familiar orange-red salt (NH4)2[Ce(NO3)6],widely used as an oxidizing agent in organic chemistry and in titrimetric analysis.

Cerium(IV) halide complexes include (NH4)4[CeF8] (square antiprismatic) and(NH4)3CeF7(H2O) (dimeric, with dodecahedral coordination in the Ce2F12

6− ions) whichare prepared by crystallization; heating the former decomposes it to (NH4)2CeF6, whichcontains infinite (CeF2−

6 )∞ chains. Heating an MCl/CeO2 mixture in fluorine gives alkalimetal salts M2CeF6 and M3CeF7 (M = Na–Cs); unlike the ammonium salts, these are water-sensitive and cannot be made by wet methods. The ammonium salts may be more stablebecause hydrogen bonding between the ammonium ions and fluoride ions could contributeto a higher lattice energy.

Although the binary chloride does not exist, a number of hexachloro salts have beenmade:

CeO2.xH2OSOCl2−−−−→

diglyme(H2 diglyme3)++ CeCl6

−−

CeO2.xH2OHCl(g)

−−−−−−−→EtOH/RCl

R2CeCl6 (R = Ph3PH, Et4N, Ph4As)

These may be converted into the analogous violet, moisture-sensitive salts of the [CeBr6]2−

ion on treatment with gaseous HBr. No salts of the [CeI6]2− ion have been reported, presum-ably for redox reasons, though a compound with a Ce(IV)–I bond has been made (Figure.4.13).

R= SiMe2But

RN

RN Ce

N

NR

IR

N

RN Ce

N

NR

I2

Figure 4.13A compound with a Ce(lr)-I bond.

The four nitrogen atoms are good donors and reduce the effective charge on the cerium,affecting its reduction potential.

[Ce{N(SiMe3)2}3] is an obvious choice of starting material for the synthesis of other CeIV

amides, but oxidation with Cl2 or Br2 was not successful; however, use of TeCl4 oxidantled to [Ce(Cl){N(SiMe3)2}3] and similarly Ph3PBr2 afforded [Ce(Br){N(SiMe3}2)3] (Fig-ure 4.14). The corresponding iodo compound has not (yet) been synthesized.

A number of neutral complexes of O-donor Lewis bases have been made, such asten-coordinate trans-Ce(NO3)4(Ph3PO)2 and octahedral CeCl4L2 [L = Ph3 PO, Ph3AsO,

...



SiMe3

N

SiMe3

SiMe3

Me3Si

N SiMe3

Ce

Me3Si

N

X

Figure 4.14X = Br, Cl

(Me2N)3PO, But2SO, (H2N)2CO]. They are prepared thus:

(NH4)2Ce(NO3)6Ph3PO−−−→Me2CO

Ce(NO3)4(Ph3PO)2 (trans)

(NH4)2Ce(NO3)6

HCl(g)−−−−→Ph3PO

CeCl4(Ph3PO)2 (cis)

H2(diglyme)3CeCl6L−−−−→

EtOAcCeCl4L2 [L = Ph3AsO; But

2SO; (Me2N)3PO; trans]

The CeCl4(Ph3PO)2 complex is cis, like the uranium analogue, whilst an X-ray diffrac-tion study of CeCl4[(Me2N)3PO]2 (trans) showed a close structural correspondence to thecorresponding UIV compound.

Ce(R1.CO.CH.CO.R2)3 are readily oxidized (O2) to Ce(R1.CO.CH.CO.R2)4, such asCe(acac)4(R1 = R2 = Me), Ce(dbm)4(R1 = R2 = Ph), and Ce(tmhd)4 (R1 = R2 = Me3C),generally found to have square antiprismatic structures, though Ce(tmhd)4 is closer tododecahedral. These are volatile dark red solids that are soluble in solvents such as benzeneand chloroform; they are volatile, with vapour pressures high enough for Metal OrganicChemical Vapour Deposition (MOCVD) use, whilst they have also been studied as possiblealternatives to lead compounds for petrol additives.

Another feature of the chemistry of cerium (IV) is the alkoxides.

(pyH)2CeCl6 + 4 ROH + 6 NH3 → Ce(OR)4 + 2 py + 6 NH4Cl

(NH4)2Ce(NO3)6 + 4 ROH + 6 NaOMe → Ce(OR)4 + 2 NH3 + 6 NaNO3 + 6 MeOH

Most are involatile oligomers, though the isopropoxide is volatile in vacuo below 200 ◦C {itsbis(propan-2-ol) adduct [Ce2(OPri)8(PriOH)2] is a dimer with 6-coordinate cerium} and itcan achieve coordinative saturation in adducts like [Ce(OR)4(thf)2] (R = CMe3, SiPh3) and[Ce(OCMe3)6]2−. Heating [Ce2(OPri)8(PriOH)2] gives [Ce4O(OPri)14], which is [Ce4(µ4

-O)(µ3-OPri)2(µ-OPri)8(OPri)4], an oxo-centred cluster similar to those found with LnIII

alkoxides.Cerium also forms genuine bis(porphyrin) and bis(phthalocyanine) complexes with the

metal in the ( +4) oxidation state.Although other (+4) oxides exist in the form of PrO2 and TbO2, no other (+4) aqua ions

are known; these oxides oxidize acids on dissolution, forming the M3+ (aq) ions. Alkalimetal fluoro-complexes Cs3LnF7 (Ln = Pr, Nd, Tb, Dy) and M2LnF6 (M = Na–Cs, Ln =Tb, Pr) are known for four other lanthanides; as with CeIV, they have to be made by drymethods, by fluorination of Cs3LnCl6 using XeF2 or MCl/Ln2O3 mixtures using F2.

...



Question 4.1 Study the data in Tables 4.1 and 4.2 (above). Plot a graph of log β againstatomic number for complex formation with EDTA4−. Deduce and explain the patterns inbehaviour for the lanthanides; compare the values of the stability constants for EDTA andfluoride complexes.Answer 4.1 The values for Lu3+ are greater than those for La3+ as the smaller Lu3+

ion has a greater charge density and stronger electrostatic attraction for a ligand. Valuesfor multidentate ligands are greater than those for monodentate ligands, partly because ofentropy factors and also because once one end of a ligand is attached, there is a higherchance of the other donor atoms attaching themselves.

Question 4.2 Suggest why the acetates of the lanthanides do not adopt structures with M3Ocores as found for many carboxylates of trivalent transition metals (e.g. Cr, Mn, Fe).Answer 4.2 A Molecular Orbital scheme for species such as [M3O(OAc)6]+ has d–p π -bonding involving the metals and the central oxygen atom. Lanthanide f orbitals are buriedtoo deeply to permit such overlap.

Question 4.3 A summary of structure types in the lanthanide chloride complexes of THFis given in Table 4.3 (data based upon G.B. Deacon, T. Feng, P.C. Junk, B.W. Skelton, A.N.Sobolev and A.H. White, Aust. J. Chem., 1998, 51, 75).

Use the list of structure types to work out the coordination number of the lanthanidefor each compound. Plot a scatter graph of Average coordination number (y axis) againstAtomic number (x axis) and comment.Answer 4.3 There is a decrease in CN as the ionic radius decreases. The isolation ofa number of complexes for some of the metal ions may reflect other factors such as thesolubility of the complexes and the stoichiometry of the reaction mixtures.

Question 4.4 Why do Ln(acac)3 form Lewis base adducts, when transition metal analoguesM(acac)3 (M, e.g., Cr, Fe, Co) do not?Answer 4.4 The lanthanides are larger than transition metals and therefore Ln(acac)3 arecoordinatively unsaturated.

Question 4.5 Why is a nine-coordinate Eu(dpm)3(terpy) known and not (for example)Eu(dpm)3(H2O)3?Answer 4.5 In terpyridyl, the three donor atoms are part of one molecule and take up lessspace than three individual molecules, even small ones.

Table 4.3 Lanthanide chloride complexes with THF

Structure types characterizedLa Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu1 2 2 2,3 3 3 3,4 4 4 4 4 4 5,6 6

Description of types:

1. La(thf)2(µ-Cl)3Ln(thf)2(µ-Cl)3La. . . .2. LaCl(thf)2(µ-Cl)2LnCl(thf)2(µ-Cl)2La . . . .3. [LnCl3(thf)4].4. [LnCl2(thf)5]+ [LnCl4(thf)2]−.5. [Cl2(thf)2Ln(µ-Cl)2Ln(thf)2Cl2].6. [LnCl3(thf)3].

...



Question 4.6 Chinese workers have reported compounds with the formula Ln(acac)3

(Ph3PO)3 (Ln, e.g., Nd, Eu, Ho). Comment on this.Answer 4.6 Even though acac is the least bulky β-diketonate ligand, it would be surprisingif all the Ph3PO groups were coordinated (recall that Ph3PO is sterically more demandingthan water as a ligand). Ultimately, only X-ray diffraction studies can settle this.

Question 4.7 Why in general does nitrate coordinate to the metal in crown ether complexesbut chloride does not?Answer 4.7 Nitrate has a small bite angle and takes up little more space in the coordinationsphere than a chloride ion; forming two Ln–O bonds is energetically more favourable thanforming one Ln–Cl bond and can outweigh the considerable Ln3+–OH2 bond energy.

Question 4.8 Lanthanide thiocyanate complexes are N-bonded, not S-bonded. Why is thisexpected?Answer 4.8 Ln3+ ions are ‘hard’ acids, which prefer to bond to ‘hard’ bases. Sulfur is a‘softer’ base than nitrogen.

Question 4.9 Explain why europium nitrate forms a 1:1 complex [Eu(terpy)(NO3)3(H2O)]but europium perchlorate forms a 3:1 complex [Eu(terpy)3](ClO4)3.Answer 4.9 Perchlorate is a weakly coordinating anion, so does not compete with ter-pyridyl; nitrate has reasonably strong coordinating tendencies, usually as a bidentate ligand,and does compete with terpy. Three nitrate groups will occupy six coordinating positions,leaving room for only one terpy.

Question 4.10 Why might Ce(R1.CO.CH.CO.R2)4 be expected to be a better MOCVDprecursor and petrol additive than Ce(R1.CO.CH.CO.R2)3?Answer 4.10 LnIII β-diketonates are often coordinatively unsaturated (Section 4.3.4) andtend to polymerize. The molecular nature of the CeIV compounds should confer greatervolatility, as well as solubility in nonpolar solvents (e.g. petrol). The CeIV oxidation statemay also assist the oxidation process.

Question 4.11 Suggest coordination numbers for the metal in each of the following:

[La(NO3)5(OH2)2]2−; [Pr(NO3)3(thf)4]; [PrCl3(15-crown-5)].

Answer 4.11 12, 10, 8 respectively.

Question 4.12 Suggest geometries for (a) [Eu[(N(SiMe3)2]3]−; (b) [Yb[(N(SiMe3)2]2

(Me2PCH2CH2PMe2)]; (c) Ln[(N(SiMe3)2]3(NCMe)2; (d) (Me4N)3[Ln(NCS)7]; (e)[Li(tmed)2]+ [LuBut

4]−.Answer 4.12 (a) Trigonal/triangular; (b) tetrahedral; (c) trigonal bipyramid with axialnitriles; (d) capped octahedron/trigonal prism; (e) tetrahedral.

Question 4.13 Comment on the oxidation state of [Ce4O(OPri)14].Answer 4.13 Taking O as –2 and OPri as –1 leads to an assignment of the oxidation stateof Ce as +4, further illustration of the stability of this state, when a redox reaction mighthave been expected.

...



Question 4.14 Ln[N(SiMe3)2]3 are pentane-soluble and sublime in vacuum at 100 ◦C,whereas the much lighter LnCl3 have high melting and boiling points and do not dissolvein organic solvents. Explain why.Answer 4.14 LnCl3 have giant ionic lattice structures, because of the relatively small sizeof the chloride ion. The nonpolar pentane cannot break up the ionic lattice to dissolve it.In contrast, the bulky silylamide ligands shield the metal ion, preventing oligomerizationand also presenting an ‘organic exterior’ to the solvent pentane. The presence of weak,essentially van der Waals’-type intermolecular forces means that little energy is needed tovapourize the amides and little energy change occurs on dissolution in a nonpolar solvent.

Question 4.15 Three routes for the synthesis of alkoxides involve (i) the reaction of LnCl3with an alkali metal alkoxide, (ii) reaction of an alkali metal with an alcohol, (iii) thereaction of a silylamide [Ln{N(SiMe3)2}3] with an alcohol. What are the advantages anddisadvantages of each?Answer 4.15 (i) The chemicals are readily available. The chloride is fairly insoluble inorganic solvents and it is possible that chlorine is retained in the product. (ii) There is noneed to make the alkali metal alkoxide, as the alcohol is the starting material. The metal mayneed careful cleaning and there may be the need for heating and a catalyst. One product is agas. (iii) The lanthanide silylamide has to be prepared first. Because of the bulky nature ofthe ligand, it may be inert to substitution, but there are no problems with chloride retentionand the reaction should be clean.

Question 4.16 Lanthanide alkoxide clusters do not contain metal–metal bonds. Transitionmetal carbonyl clusters frequently do. Comment.Answer 4.16 Transition metal carbonyls have transition metals in low oxidation stateswhere d orbitals can overlap well. f orbital involvement in bonding in lanthanide compoundsis minimal and the high oxidation state will contract the orbitals even further.

Question 4.17 Given the Zn2+/Zn reduction potential of −0.76.V, and using values forLn3+/Ln2+ given in Table 2.7, explain why this can be used to separate Eu from otherlanthanides that form Ln2+ ions.Answer 4.17 Since E = −0.76 V for Zn2+ +2 e− → Zn, it will reduce Eu3+ to Eu2+; sincethe reduction potential for Eu3++ e− → Eu2+ is −0.34 V, the potential for the reaction 2Eu3++ Zn →2 Eu2++ Zn2+ is + 0.42 V, and the reduction is feasible. However, thereduction potentials for the other Ln3+ ions are all more negative than that for Zn2++ 2e− → Zn; for the next most easily reduced lanthanide (III) ion, ytterbium, E = −1.05 Vfor Yb3++ e− → Yb2+, and thus zinc will not reduce Yb3+ to Yb2+.

SPH SPH


5 Electronic and Magnetic Propertiesof the Lanthanides


� know how to use Hund’s rules to work out the ground state of lanthanide ions;� calculate the magnetic moments for these ions, given the appropriate formula;� understand why spin-only values for the moments are inadequate;� state why the magnetic moments measured for Sm3+ and Eu3+ compounds differ from

these predictions;� appreciate why electronic (excitation) spectra show little dependence on compound, for

a particular metal ion;� be aware of the existence of hypersensitive transitions;� interpret electronic spectra of compounds in terms of energy level diagrams;� understand the causes of fluorescence in compounds of the lanthanides, particularly

compounds of Eu3+ and Tb3+;� interpret the fluorescence spectra of Eu3+ compounds in terms of the site symmetry;� understand the principles involved in applications, including Nd3+ in lasers, and the use

of lanthanides in lighting and TV tubes;� explain the use of lanthanide complexes as MRI agents and NMR shift reagents.

5.1 Magnetic and Spectroscopic Properties of the Ln3+ Ions

These may be accounted for using the Russell–Saunders coupling scheme inwhich the elec-tron spins are coupled together separately from the coupling of theorbital angular momentaof the electrons, and the orbital moment is unquenched. The ground state for a given lan-thanide ion is unaffected by the ligands bound to it – and thus crystal field splittings areweak – because of the shielding of the 4f electrons by the filled 5s and 5p orbitals.

The spins of the individual electrons(s) are coupled together (added vectorially) to givethe spin quantum number for the ion (S). The orbital angular momenta (l) of the individualelectrons are coupled similarly.

For an f electron, l = 3, so that the magnetic quantum number ml can have any one ofthe seven integral values between +3 and −3. Vectorial addition of the ml-values for the felectrons for the multi-electron ion affords L , the total orbital angular momentum quantumnumber:

There is a weaker coupling, spin–orbit coupling, between S and L .


SPH SPH


62 Electronic and Magnetic Properties of the Lanthanides

State

L 0 1 2 3 4 5 6 7

S P D F G H I Ksymbol

Figure 5.1State symbols for different values of L.

ml 3 0 −1 −2 −3

↑ ↑ ↑ ↑ ↑

2 1

Figure 5.2Box diagram for Sm3+.

Vector addition of L and S affords the resulting quantum number, J . J can have values of(L + S), (L + S) − 1; . . . . . . . . (L − S). To take an example, if L = 6 and S = 2, J -valuesof 8, 7, 6, 5, and 4 are possible.

For any ion, a number of electronic states are possible. The ground state can be determinedusing Hund’s rules (in this order):

1. The spin multiplicity (2S + 1) is as high as possible.2. If there is more than one term with the same spin multiplicity, the term with the highest

L-value is the ground state.3. For a shell less than half-filled, J for the ground state takes the lowest possible value;

for a shell more than half-filled, J for the ground state is the highest possible.

To give an example, working out the term symbol for the ground state of Sm3+ (f 5) Firstcomplete a ‘box diagram’, representing orbitals by boxes (7 boxes for 7 f orbitals) andelectrons by arrows (Figure 5.2). Put electrons in separate orbitals when the shell is lessthan half-filled, i.e. choosing the maximum number of unpaired electrons, and choosing tomaximize the values of ml to give the highest L-value (check Hund’s rule 2).

So S = ∑ms = 5/2, therefore 2S + 1 = 2(5/2) + 1 = 6.

L = ∑m1 = +3 + 2 + 1 + 0 − 1 = +5, so it is an H state.

J can have the values of (L + S); (L + S) − 1; (L + S) − 2; . . . . . . .; (L − S), so here J =(5 + 5/2); (5 + 5/2) − 1; (5 + 5/2) − 2 . . . . . . . (5 − 5/2) = 15/2; 13/2; 11/2; 9/2; 7/2;5/2.

Since the shell is less than half-filled, the state with the lowest J -value is the ground state(Hund’s third rule), so this is J = 5/2.

The term symbol for the ground state of Sm3+ is thus 6H5/2

5.2 Magnetic Properties of the Ln3+ Ions

With the exception of La3+ and Lu3+ (and of course Y3+) the Ln3+ ions all contain unpairedelectrons and are paramagnetic. Their magnetic properties are determined entirely by theground state (with two exceptions we shall encounter), as the excited states are so wellseparated from the ground state (owing to spin–orbit coupling; see Figure 5.3) and are thusthermally inaccessible.

SPH SPH


Fig

ure

5.3

Ene

rgy

leve

lsof

the

Ln3+

ions

[fro

mS.

Huf

ner,

inS.

P.Si

nha

(ed.

),Sy

stem

atic

san

dP

rope

rtie

sof

the

Lan

than

ides

,R

eide

l,D

ordr

echt

,19

83;

repr

oduc

edby

perm

issi

onof

the

publ

ishe

ran

dau

thor

].

63

SPH SPH



The magnetic moment of the Ln3+ ions is essentially independent of environment, sothat one cannot distinguish between coordination geometries as is sometimes possible fortransition metals – in the case of octahedral and tetrahedral Co2+ complexes, for example.The moments are given by the equation:

µeff = gJ√

J (J + 1)

where the Lande g-factor is defined by:

gJ = [S(S + 1) − L(L + 1) + 3J (J + 1)]/2J (J + 1); this may also be written

gJ = 3/2 + [S(S + 1) − L(L + 1)]/2J (J + 1)

This differs from the familiar µeff = √n(n + 2) formula, where n is the number of unpaired

electrons, [or√

4S(S + 1), spin-only formula], often applicable to the 3dn transition metalions, as in the latter case the orbital contribution to the moment is quenched by the interactionof the metal’s 3d orbitals with the ligands.

The magnetic moments in the second half of the series are greater than the moments inthe first half, as J = L + S for a shell greater than half-filled and J = L − S for a less thanhalf-filled shell.

Example: Calculate the magnetic moment for a complex of Ho3+, such asHo(phen)2(NO3)3.

The ground state is 5I8 (2S + 1L J ), since 2S + 1 = 5, S = 2; L = 6 (I state); J = 8. gJ

must first be calculated; gJ = 3/2 + [S(S + 1) − L(L + 1)]/2J (J + 1)

Substituting, gJ = 3/2 + [2(2 + 1) − 6(6 + 1)]/2 × 8(8 + 1) = 3/2 − 36/144, so gJ = 5/4.

Now substitute in µeff = gJ√

J (J + 1); µeff = 5/4√

8(8 + 1) = 10.60 µB

The strength of spin–orbit coupling means that the ground state is well separated fromexcited states, except for Sm3+ and Eu3+, where contributions from low-lying paramagneticexcited states (see Figure 5.3) contribute to the magnetic moment. Thus, if the magneticproperties of the Eu3+ ion were solely determined by the 7F0 ground state, its compoundswould be diamagnetic, whereas contributions from thermally accessible levels such as 7F1

and 7F2 [using a Boltzmann factor of exp(−�E/kT )] lead to the observed room-temperaturemagnetic moments in the region of 3.5 µB. Similarly, in the case of Sm3+, thermal populationof the 6H7/2 state leads to moments around 1.6 µB, rather than the value of 0.845 µB thatwould be expected if just the 6H5/2 ground state were responsible.

5.2.1 Adiabatic Demagnetization

This application of the high magnetic moments of lanthanide complexes is a process bywhich the removal of a magnetic field from certain materials lowers their temperature; itwas proposed independently in 1926–1927 by Peter Debye and William Francis Giauque(Giauque won the Nobel prize for Chemistry in 1949 for his low-temperature work). It isapplied particularly to cooling an extremely cold material at ∼1 K to much lower tempera-tures, perhaps as low as 0.0015 K. An Adiabatic Demagnetization Refrigerator uses certainhighly paramagnetic lanthanide salts [e.g. Gd2(SO4)3.8H2O] immersed in a bath of liquidhelium; the lanthanide ions in the second half of the series have higher magnetic moments(see Table 5.1) so compounds of Gd3+ and Dy3+ tend to be used.

SPH SPH


Energy Level Diagrams for the Lanthanide Ions, and their Electronic Spectra 65

Table 5.1 Magnetic Moments of Ln3+ ions at room temperature

f n Ground term Predicted µeff (µB) µeff M(phen)2(NO3)3 (µB)

La 0 1S0 0.00 0Ce 1 2F5/2 2.54 2.46Pr 2 3H4 3.58 3.48Nd 3 4I9/2 3.68 3.44Pm 4 5I4 2.83Sm 5 6H5/2 0.85 1.64Eu 6 7F0 0.00 3.36Gd 7 8S7/2 7.94 7.97Tb 8 7F6 9.72 9.81Dy 9 6H15/2 10.63 10.6Ho 10 5I8 10.60 10.7Er 11 4I15/2 9.59 9.46Tm 12 3H6 7.57 7.51Yb 13 2F7/2 4.53 4.47Lu 14 1S0 0.00 0

If the paramagnetic substance is placed in a strong external magnetic field, the magneticdipoles tend to align with the field. This is an exothermic process, as ‘aligned’ is the lowestenergy state, and unaligned dipoles need to lose energy to achieve the lowest state. Theheat energy released is removed by the external coolant, and the magnetic material (and itssurroundings) cool back to their original temperature (say 1 K). Thermal contact betweenthe material and the liquid helium bath is now broken (a ‘heat switch’ is turned off) and themagnetic field is turned off. The magnetic dipoles tend to randomize now, absorbing thermalenergy from the magnetic material, which cools (along with its immediate surroundings),producing the cooling in the sample.

5.3 Energy Level Diagrams for the Lanthanide Ions, and their Electronic Spectra

Energy level diagrams for all the Ln3+ ions are shown in Figure 5.3.These are based on theoretical predictions, coupled with experimental results. Accurate

values have been determined in many experimental situations, such as halide lattices likeLnF3, which closely resemble those of the gaseous ions. In the lanthanide ions, the filled‘outer’ 5s and 5p orbitals efficiently shield the 4f electrons from surrounding ligands, withthe result that crystal field splittings are of the order of 100 cm−1. The weak crystal fieldsplittings can thus be treated as a perturbation upon the free-ion levels; in contrast, of course,in the 3d metals, CF splittings are large and L–S coupling is weak. A consequence of thelack of CF effects in the lanthanides is that thermal motion of the ligands has very littleeffect upon them (as is not the case in the 3d situation), so the f–f absorption bands in thespectra are very narrow, almost as narrow as for free (gaseous) ions.

5.3.1 Electronic Spectra

Most lanthanide ions absorb electromagnetic radiation, particularly in the visible regionof the spectrum, exciting the ion from its ground state to a higher electronic state, as aconsequence of the partly filled 4f subshell. The f–f transitions are excited both by magneticdipole and electric dipole radiation. Normally the magnetic dipole transitions would not

SPH SPH



be seen, but in the case of the lanthanides the electric dipole transitions are much weakerthan for transition metal complexes and magnetic dipole transitions can often be seen,especially in fluorescence spectra. The magnetic dipole transitions are parity-allowed, whilstthe electric dipole transitions are parity-forbidden (‘Laporte-forbidden’) in the same sense asd–d transitions in transition metal ions. The f–f transitions gain intensity through mixing inhigher electronic states (including d states) of opposite parity, either through the (permanent)effects of a low-symmetry ligand field or through asymmetric molecular vibrations thatmomentarily destroy any centre of symmetry – ‘vibronic coupling’ – though the effect isstill weaker than in transition metal complexes.

Not all the lanthanide ions give rise to f–f transitions, including obviously the f 0 and f 14

species, La3+ and Lu3+. Likewise there are no f–f transitions for the f 1(Ce3+) and f 13 (Yb3+)ions, as with only a single L-value there is no upper 4f state. Transitions between 2F5/2 and2F7/2 are seen in the case of Ce3+ as a rather broad band in the infrared region around2000 cm−1. Ce3+ and Yb3+do, however, give rise to broad 4f n → 4f n−15d1 transitions(as indeed do many lanthanides). Even an ion like Eu3+, which has several absorptions inthe visible region of the spectrum, has only weak absorptions, so many of its compoundsappear colourless; the only three tripositive ions whose compounds are invariably colouredare Pr3+, Nd3+, and Er3+.

Colours of the Ln3+ ions in aqueous solution are listed in Table 5.2. A typical spectrumis that of Pr3+, shown in Figure 5.4; note the sharp absorption bands, with extinctioncoefficients below unity.

The electronic spectra of lanthanide compounds resemble those of the free ions, in contrastto the norm in transition metal chemistry; the crystal-field splittings can be treated as aperturbation on the unsplit 2S + 1L J levels. Complexes thus have much the same colour as thecorresponding aqua ions. Thus one cannot distinguish between coordination geometries aswith octahedrally and tetrahedrally coordinated Co2+, for example (let alone note profoundcolour differences). As already noted, a further consequence of the weak CF splittings isthe sharpness of the f–f transitions. The spectrum of the complex Eu(NO3)3(15-crown-5)shown in Figure 5.5 is another example of this.

The previous discussion has centred upon the Ln3+ ions. Most Ln2+ ions are not stablein solution, but have been prepared artificially in lattices by doping CaF2 with the Ln3+

Table 5.2 Colours of Ln3+ ions in aqueous solution

f n Colour

La3+ f 0 ColourlessCe3+ f 1 ColourlessPr3+ f 2 GreenNd3+ f 3 VioletPm3+ f 4 PinkSm3+ f 5 Pale yellowEu3+ f 6 ColourlessGd3+ f 7 ColourlessTb3+ f 8 V.pale pinkDy3+ f 9 Pale yellowHo3+ f 10 YellowEr3+ f 11 RoseTm3+ f 12 Pale greenYb3+ f 13 ColourlessLu3+ f 14 Colourless

SPH SPH


0.06

0.04

0.02

Mol

ar a

bsor

ptio

n co

effic

ient

(ε)

400 450Wavelength (nm)

500 550

1D2

3P0

3P1

3P2

Figure 5.4Electronic absorption bands in the spectrum of PrCl3 (aq) reproduced with permission from S.A.Cotton, Lanthanides and Actinides, Macmillan (1991) p. 30.

Figure 5.5(a) Excitation spectrum at 77 K of Eu(NO3)3·15-crown-5 (λanal = 618 nm). (b) Fluorescence spectrumat 77 K (λexc = 397.7 nm) (reproduced by permission of the American Chemical Society from J.-C.G.Bunzli, B. Klein, G. Chapuis, and K.J. Schenk, Inorg. Chem., 1982, 21, 808).

67

SPH SPH



ions, then γ -irradiating. The reduced charge of the Ln2+ ion causes the energy levels to becloser together, so the ions have different colours; thus whilst Gd3+ ions are colourless, theisoelectronic (f 7) Eu2+ ions are yellow.

5.3.2 Hypersensitive Transitions

In contrast to the situation with the 3d transition metals in particular, the 4f–4f transitions inthe electronic spectra of lanthanide complexes rarely serve any diagnostic purpose. It maybe noted, however, that the spectra of the octahedral [LnX6]3− ions (X = Cl, Br) have par-ticularly small extinction coefficients, an order of magnitude lower than the correspondingaqua ions, due to the high symmetry of the environment.

Some transitions are ‘hypersensitive’ to changes in the symmetry and strength of theligand field; as a result, they display shifts of the absorption bands, usually to longer wave-length, as well as band splitting and intensity variation. It is most marked for Ho3+, Er3+

and particularly for the 4I9/2 →2H9/2, 4F5/2 and 4I9/2,→4G5/2, 4G7/2 transitions in the caseof the Nd3+ ion.

Figure 5.6 demonstrates how the profile of the band caused by the 4I9/2 →2H9/2, 4F5/2

transition in the spectrum of Nd3+ (aq) resembles that of the tricapped trigonal prismatic

(a)

(b)

(c)

(d)

(e)

8200

λ (Å)

80007800

Abs

orba

nce

Figure 5.6Spectra of the Nd3+ 4I9/2 → 2H9/2, 4F5/2 transitions: (a) solid Nd(BrO3)3·9H2O; (b) 5.35 × 10−2MNd3+ (aq); (c) 5.35 × 10−2M Nd3+ in 11.4M HCl; (d) solid NdCl3·6H2O; (e) Solid Nd2(SO4)3·8H2O(reproduced by permission of the American Chemical Society from D.G. Karraker, Inorg. Chem.,1968, 7, 473).

SPH SPH


Luminescence Spectra 69

[Nd(H2O)9]3+ ions in Nd(BrO3)3.9H2O, but is significantly different from that of the 8-coordinate Nd3+ in the hydrated sulfate and chloride salts. This was important evidence infavour of the aqua ion being 9 coordinate. The spectrum of Nd3+ in concentrated HCl issignificantly different, and suggests that an 8-coordinate species is present here.

5.4 Luminescence Spectra

Many lanthanide ions exhibit luminescence, emitting radiation from an excited electronicstate, the emitted light having sharp lines characteristic of f–f transitions of a Ln3+ ion. Aswill be seen later, this can be enhanced considerably by attaching a suitable organic ligand(e.g. a β-diketonate, phenanthroline, crown ether, etc.) to the lanthanide.

The process occurs as summarized in Figure 5.7.The mechanism of luminescence of a lanthanide complex is as follows. An electron is

promoted to an excited singlet state in a ligand upon absorption of a quantum of energy(e.g. from ultraviolet light). This photon drops back to the lowest state of the excited singlet,from where it can return to the ground state directly (ligand fluorescence) or follow a non-radiative path to a triplet state of the ligand. Thence it may either return to the ground state(phosphorescence) or alternatively undergo non-radiative intersystem crossing, this timeto a nearby excited state of a Ln3+ ion, whence it can return to the ground state either bynon-radiative emission or by metal-ion fluorescence involving an f–f transition.

Ligand phosphorescence

Ligand fluorescence

Lanthanide fluorescence

Excited triplet

Excited singlet

Ln3+

excited states

Radiative absorption

Radiative emission (fluorescence)

Ground- state singlet

Nonradiative emission

Figure 5.7Luminescence in lanthanide complexes.

SPH SPH



Certain Ln3+ ions have excited states lying slightly lower in energy than the triplet statesof typical ligands, and exhibit strong metal-ion fluorescence, most markedly for Eu3+ andTb3+, and possibly for others such as Sm3+ and Dy3+. Among the other Ln3+ ions, La3+

and Lu3+ have no f n excited state; Gd3+ has all its excited states above the ligand tripletstates while the others have a large number of excited states promoting energy loss by anon-radiative route. Tb3+ and Eu3+ are thus the two most useful ions for these studies. Theyluminesce with green and red colours, respectively. For Tb3+ the main emissions responsibleare 5D4 →7Fn(n = 6–0) with 5D4 → 7F5 the strongest, whilst for Eu3+, 5D0 →7Fn areseen (n = 4–0) with the transitions to 7F0, 7F1, and 7F2 the most useful. The process isimportant in lighting applications, and also in both qualitative and quantitative analysis,and in luminescent imaging.

Although the energy level separations are little affected by ligand, nevertheless they arenot the same in all complexes. Careful study of the 7F0 →5D0 separation in a number ofEu3+ complexes has shown its shift towards lower energies to depend upon the donor atomsinvolved, the actual values of the separation varying between 17 232 cm−1 in [Eu(dpa)3]3−

(dpa = dipicolinate) and 17 280 cm−1 in [Eu(H2O)9]3+ in some 30 complexes studied,reflecting changes in covalency in the bond.

The ligand field in a complex ion removes the degeneracy of a given 2S+1 LJ term partlyor completely, the extent of this and the resultant splitting of the emission line dependingupon the symmetry of the ligand field (Table 5.3). This means that, in many cases, theindividual transitions in the luminescence spectra consist of more than one line (subject toresolution).

Study of the intensity and splitting pattern of certain transitions in the fluorescence spectraof compounds of Eu3+ and Tb3+ in particular can give a good deal of information aboutthe environment of the lanthanide ion. The rules are summarized in Tables 5.4 and 5.5for both europium and terbium. Owing to the weakness of the electric dipole transitions,the magnetic dipole transitions are of comparable intensity. (For Eu3+, 5D0 →7Fx arepredominantly electric dipole for x even, magnetic dipole for x odd).

Table 5.3 Number and degeneracy of 5D0–7FJ transitions of Eu3+ ions in some common symmetries

7F07F1

7F27F3

7F4

Symmetry ED MD ED ED ED

Ih none T1g none none noneOh none T1g none T1g noneTd none T1 T2 T1 T2

D4h none A2g + Eg Eg A2g + Eg noneD4d none A2 + E3 E1 A2 + E3 B2 + E1

D2d none A2 + E B2 + E B2 + 2 E B2 + 2ED3h none A′

2 + E′′ E′ A′2 + E′′ A′′

2 + 2E′

D3d none A2g + Eg A1g + Eg none noneD3 none A2 + E 2E 2A2 + 2E A2 + 3EC3v A1 A2 + E A1 + 2E A1 + 2E 2A1 + 3EC3 A A + E A + 2E 3A + 2E 3A + 3EC2v A1 A2 + B1 + B2 2A1 + B1 + B2 A1 + 2B1 + 2B2 3A1 + 2B1 + 2B2

C2 A A + 2B 3A + 2B 3A + 4B 5A + 4BCi A 3A 5A 7A 9ACs A′ A′ + 2A′′ 3A′ + 2A′′ 3A′ + 4A′′ 5A′ + 4A′′

SPH SPH



Table 5.4 Features of 5D0 → 7FJ luminescent transitions for Eu3+

J Main character Region (nm) Intensity Comments

0 ED 577–581 V.weak Absent in high symmetry – ‘forbidden’1 MD 585–600 Strong Intensity largely independent of environment2 ED 610–625 V.weak to v.strong Absent if ion on inversion centre; hypersensitive3 ED 640–655 V.weak ‘Forbidden’4 ED 680–710 Medium to strong Environment-sensitive

Table 5.5 Features of 5D4 → 7FJ luminescent transitions for Tb3+

J Region (nm) Intensity Comments

6 480–505 Medium to strong Environment-sensitive5 535–555 Strong to v.strong Good probe4 580–600 Medium to strong Environment-sensitive3 615–625 Medium2 640–655 Weak Environment-sensitive

As the luminescence technique is one that can be applied particularly to lanthanidecomplexes we will discuss it in some depth; and since interpretation is simplest in the case ofEu3+, so discussion will centre on that, particularly since it is most involved in applications.In the case of the europium(III) ion, the 7F0 ground state is unsplit, so that transitions to it givestraightforward information about the excited state. Because the 5D0 state is also unsplit, ifmore than one component is seen for this transition, it shows more than one europium site.This can be seen in the luminescence spectrum of the crown-ether complex [Eu(NO3)3(15-crown-5)] shown in Figure 5.5, which gives some useful information. This compoundcontains a pentadentate macrocycle and three bidentate nitrates, leading to 11-coordinateeuropium. A single, very sharp line is seen for the 5D0 →7F0 transition, indicating onlyone europium site. Secondly, the fact that this electric-dipole transition is seen indicatesthat the site symmetry is not especially high (i.e. it is not on an inversion centre). The5D0 →7F1 transition (magnetic-dipole allowed and relatively insensitive to environment)appears as a doublet and singlet, a splitting consistent with the approximately pentagonalpseudosymmetry seen in its crystal structure (in a lower-symmetry environment, the doubletwould be further split). The 5D0 →7F2 transition is electric dipole in origin; again, it is absentif the ion is on an inversion centre, and its intensity is very sensitive to environment, makingit a good ‘probe’, so its strength here indicates a relatively low symmetry of the site.

An important example historically of the application of luminescence spectroscopy(1969) is [Eu(terpy)3] (ClO4)3, whose spectrum (Figure 5.8) supported the assigned geome-try before crystallographic details were available (crystallographic structural determinationwas a much slower business at that time). The 5D0 →7F0 transition is absent, whilst the5D0 →7F1 transition consists of a singlet and a very slightly split (0.4 nm) doublet. The5D0 →7F2 transition is two split doublets, whilst the 5D0 →7F4 transition is composed ofthree slightly split doublets and a singlet. If the symmetry were perfectly D3, none of thedoublets would be split (check Table 5.3), so these splittings can be explained by a veryslight descent of symmetry below D3.

The effect of site symmetry upon luminescence is not confined to structuraldetermination, it also affects lighting applications. Compare the emissions of Eu3+ in thetwo oxides NaInO2 and NaGdO2 (Figure 5.9); both have structures based on NaCl, but in

SPH SPH



Figure 5.8Fluorescence spectrum of [Eu(terpy)3]3+ ions in [Eu(terpy)3] (ClO4)3, reproduced from D.A. Durham,G.H. Frost, and F.A. Hart, J. Inorg. Nucl. Chem., 1969, 31, 833 by permission of Pergamon PressPLC.

the former the europium ions are on centrosymmetric sites, so the 5D0 →7F2 transition isabsent, and the emission is predominantly around 590 nm from the 5D0 →7F1 transition.In the case of Eu3+ in NaGdO2, the europium site lacks an inversion centre, so in thiscase the emission is largely from the strongly radiating (electric dipole allowed) 5D0 →7F2

transition at 610 nm, the difference in wavelength producing a significant shift in colour(see section 5.4.4 for more on this).

Figure 5.9Luminescence from Eu3+ substituted in NaInO2 and NaGdO2, showing the effect of site symmetryupon emission (adapted from on B. Blasse and A. Bril, J. Chem. Phys., 1966, 45, 3327).

SPH SPH



5D0

7F6

7F0

Ligandsinglet Ligand

tripletEu3+states

Intersystem crossing

Nonradiative transition

Radiative transition

νOH

Figure 5.10Quenching of luminescence of Eu3+.

5.4.1 Quenching

One problem with luminescence in aqueous solution is that another pathway is available fordeactivation of the excited state of the lanthanide, in the form of vibrational energy transferto water molecules in particular. Figure 5.10 shows this process [in addition to showingthe processes involved in EuIII luminescence (only a selection of the radiative transitionsare shown]. This ‘quenching’ of luminescence can be minimized (apart from working inthe solid state, or using non-aqueous solvents) by using multidentate ligands which excludewaters from the coordination sphere of the metal, and also by using ligands which tend toencapsulate the lanthanide ion.

5.4.2 Antenna Effects

Luminescence from lanthanides is inherently weak (Laporte-forbidden transitions). Deac-tivation by vibrational transfer can be reduced, as discussed in the last section; another waythat gives considerably enhancement is to use a chromophore as a ligand, which absorbs asuitable wavelength of radiation strongly (acts as an ‘antenna’) which it can then transfer tothe lanthanide and excite it to the emissive state, using excitation in the region 330–430 nm(Figure 5.11). The most likely acceptor levels for Eu3+ and Tb3+ are 17 200 and 20 400 cm−1

respectively, so the triplet level in the acceptor ligand needs to be above 22 000 cm−1, other-wise competing thermally activated back-energy-transfer occurs. Some of the most suitableligands that have been utilized in the past 20 years include Lehn’s tris(bipyridyl) cryptand;calixarenes; and substituted macrocycles containing phenanthridines. The acceptor chro-mophore does not need to be directly bound to the lanthanide, but should be close for bestresults. Some lanthanides emit in the near-IR, notably Yb3+, Nd3+, and Er3+; these can beexcited directly with an optical parametric oscillator.

SPH SPH



Antenna

Organic ligand

Ln3+Ln3+

Ln3+ (excited state)

Luminescent emission

Energy transferhν

Figure 5.11The ‘Antenna’ effect.

5.4.3 Applications of Luminescence to Sensory Probes

There is considerable interest in using lanthanide complexes as probes for the presence ofparticular cations and anions, oxygen, etc., with obvious potential in biological, clinical,and environmental applications. Some examples follow.

Luminescence of the Eu3+ complex of the tetradentate tris(2-pyridylmethyl)amine ligandshown in Figure 5.12 (R = Me) shows a particular sensitivity for nitrate (over other ionssuch as chloride, sulfate, and acetate), greatest enhancement of the luminescence spectrumbeing for the ‘hypersensitive’ 5D0 →7F2 transition at 618 nm, as might be expected (seeT. Yamada, S. Shinoda, and H. Tsukube, Chem. Commun., 2002, 218).

The terbium complex, in contrast, exhibits greatest sensitivity for chloride. Using theachiral ligand (R = H) exhibits similar selectivity for these anions, but with rather lesssensitivity.

A terbium complex (Figure 5.13) is a molecular logic gate corresponding to a two-inputINHIBIT function; the output (a terbium emission line) is observed only when the ‘inputs’,the presence of proteins and the absence of oxygen, are both satisfied.

A similar complex (Figure 5.14, M = Tb) exhibits a luminescence enhancement by afactor of 125, comparing the emissive neutral complex with its protonated form (pH 3), pro-tonation (of the phenanthridyl nitrogen) suppressing electron transfer. Protonation reducesthe energy of the ligand triplet state (the triplet is 1500 cm−1 higher than the terbium 5D4

state in the neutral complex, compared with only 800 cm−1 higher in the protonated form),favouring deactivation of the terbium complex by back-energy-transfer to the triplet. Con-versely, ligand-based fluorescence increases appreciably on protonation. The triplet stateof the phenanthridyl moiety is quenched by O2, so removing oxygen from a solution of thecomplex steadily increases the terbium lifetime and emission intensity. This complex thusacts as an oxygen sensor, as does the N-methylated analogue (independent of pH in therange 2–9). As strong terbium luminescence only occurs in the absence of both acid and

N

N

N

N

R

Figure 5.12A tris(2-pyridylmethyl)amine ligand.

SPH SPH



N CH3

Tb

N

CH2CH2

PCH3

O

O

NCH2

C

NH

O

N

H2C

PH3C

O

O

O

O

H3C

P N

Figure 5.13See T. Gunnlaugsson, D.A. MacDonail, and D. Parker, Chem. Commun., 2000, 93.

O2, this complex can be considered as a molecular NOR gate, with H+ and O2as inputs andthe Tb luminescence as the output.

The europium complex of this ligand (Figure 5.14, M = Eu) exhibits considerable Eu3+

luminescence enhancement on acidification, comparing the neutral complex with the pro-tonated form at pH 1.5. The N-methylated analogue exhibits europium luminescence thatis selectively quenched by chloride ions in the presence of phosphate, citrate, lactate, andbicarbonate, all potentially present in a cellular situation, showing potential as a luminescentsensor in bioassays. Such complexes have further been incorporated into sol–gel thin filmsand glasses that show good stability to leaching and photodegradation.

5.4.4 Fluorescence and TV

Colour televisions and similar displays are the largest commercial market for lanthanidephosphors, with over 100 million tubes manufactured a year. About 2 g of phosphor is usedin each tube. The lanthanide involvement in a traditional colour TV tube works along these

M

N

CH2

PCH3

O

O

NCH2

C

NH

O

N

H2C

PH3C

O

O

O

H3C

P

CH2

N

N

O

Figure 5.14(see D. Parker and J.A.G. Williams, Chem. Commun., 1998, 245; D. Parker, K. Senanaayake, andJ.A.G. Williams, Chem. Commun., 1997, 1777).

SPH SPH



lines: There are three electron guns firing electron beams at the screen from subtly differentangles (using a ‘mask’ of some kind) to hit clusters, each comprising three phosphor ‘dots’,each ‘dot’ emitting a different primary colour, the ‘mask’ aligned so the appropriate electrongun fires at its matching phosphor. For many years, the red-emitting phosphor has used aEu3+ material, originally Eu3+:YVO4, more recently Eu3+ in Y2O2S or Eu 3+:Y2O3. Theseare employed in preference to broad-band emitters like Ag:Zn,CdS as although they areenergetically less efficient, these ‘narrow-band’ lanthanide phosphors are brighter (and alsomatch the eye’s colour response better). Green light is obtained from either Cu,Al:Zn,CdSor Ce3+:CaS and Eu2+:SrGa2S4 (these are ‘broad-band’ emitters) or the narrow-lineTb3+:La2O2S. Tm3+:ZnS has been suggested for the blue phosphor but Ag,Al:ZnS is gen-erally used. Development of materials for alternative flat-screens proceeds apace.

5.4.5 Lighting Applications

There is intense interest in using rare earths in lighting applications. The so-called tricolourlamps use three narrow-band lanthanide light-emitting materials with maxima around 450,540, and 610 nm; such as Eu2+:BaMgAl10O17; (Ce,Gd,Tb)MgB5O10; and Eu:Y2O3, re-spectively. This gives a good colour rendering at high efficiency. Research continues indeveloping materials for better projection materials, flat displays, and plasma displays.Eu2+-based materials can be used for blue phosphors but tend to suffer from degradationunder vacuum UV excitation, whilst a problem with the widely used Eu3+:(Y,Gd)BO3 ‘red’phosphor is that, owing to the high symmetry of the Eu3+ site, the 5D0 →7F1 emission isvery pronounced and the colour is orange-biased (see Section 5.4); however, if the symme-try is too low, there is too much far-red and IR radiation from the 5D0 →7F4,6 transitions!Thus new materials are continually being synthesized and tested.

One unusual application is a UV emitter containing Eu2+:SrB4O7; insects can ‘see’ theUV and be attracted to it, so it has been used in insect traps.

5.4.6 Lasers

(Light Amplification by Stimulated Emission of Radiation) Various lanthanide ions can beused in lasers, different ions operating at different frequencies. The most popular, however,is the neodymium laser, most usually using Nd3+ ions in yttrium aluminium garnet (YAG;Y3Al5O12). Such a laser functions by increasing light emission by stimulating the release ofphotons from excited Nd3+ions (in this case). A typical device consists of a YAG rod a fewcm long (containing about 1% neodymium in place of yttrium) that is fitted with a mirror ateach end, one being a partly transmitting mirror (or a similar device). A tungsten-halogenlamp (or some similar device) is used to ‘pump’ the system to ensure that an excess ofNd3+ ions is in an excited state (e.g. 4F5/2 or 4F7/2) so that more ions can emit electronsthan can absorb; these excited ions decay rapidly (or ‘cascade’) to the long-lifetime 4F3/2

state non-radiatively, so that a high proportion of Nd3+ ions are in this state rather than theground state, a ‘population inversion’ (Figure 5.15).

If a photon of the correct energy (at the wavelength of the laser transition) hits a Nd3+ ion inthe 4F3/2 state, the Nd3+ ion is stimulated to release another photon of the same wavelength,as it drops to the 4I11/2 state. As the photons are reflected backwards and forwards in therod, more and more ions are stimulated into giving up photons (thus depopulating the 4F3/2

state) and eventually the build up of photons is so great that they emerge from the rod asan intense beam of coherent monochromatic light (wavelength 1.06 µm, in the near-IR).The 4I11/2 state is an excited level of the ground state, which is not thermally populated and

SPH SPH


NMR Applications 77

}

Relaxation

Excitation Lasertransition

2000 cm−1

Fastrelaxation

4S3/2, 4F7/24F5/2, 4H9/2

4I9/2

4I11/2

4F3/2

Figure 5.15A ‘four-level’ Nd3+ laser. Reproduced by permission of Macmillan from S.A. Cotton, Lanthanidesand Actinides, Macmillan, 1991, p. 32.

so undergoes rapid relaxation to the ground state, maintaining the ‘population inversion’(whereupon the laser process can recommence). Neodymium is thus said to act as a ‘four-level’ laser.

5.4.7 Euro Banknotes

These exhibit green, blue, and red luminescent bands under UV irradiation, as a secu-rity measure. The red bands are doubtless due to some Eu3+ complex, probably with aβ-diketonate or some similar ligand. As we have seen, there are Eu2+ complexes thatcould cause the green and blue luminescence. Researchers at the University of Twentein the Netherlands suggest that a likely candidate for the source of the green colour isSrGa2S4:Eu2+, and that the blue colour may be caused by (BaO)x .6Al2O3:Eu3+. It’s quiteappropriate that Euro notes contain europium, really.

5.5 NMR Applications

5.5.1 β-Diketonates as NMR Shift Reagents

Paramagnetic lanthanide β-diketonate complexes Ln(R1COCHCOR2)3 produce shifts in theNMR spectra of Lewis base molecules capable of forming adducts with them and are thusoften referred to as Lanthanide Shift Reagents (LSRs) though all paramagnetic lanthanidecomplexes can exhibit shifted resonances. Molecules were chosen, such as Eu(dpm)3 (R1 =R2 = Me3C), which were quite soluble in non - polar solvents. The magnitude of the protonshifts depends upon the distance of the proton from the site of coordination to the lanthanideion.

Immense activity in this area in the early 1970s resulted as the use of LSRs enabledsimplification of the spectra of organic molecules without the use of high-frequency spec-trometers. The spreading-out of the spectrum and differential nature of the shifts removeddegeneracies and overlaps, whilst study of the shifts, particularly when more than one LSRwas used, permitted spatial assignment of the protons (or other resonant nucleus) withconcomitant structural information about the organic molecule.

SPH SPH



δ/

Figure 5.16The 100 MHz proton NMR spectrum of benzyl alcohol in the presence of Eu(dpm)3 (0.39 mol)(reproduced by permission of the Royal Society of Chemistry from J.K. M. Sanders and D.H. Williams,Chem. Commun., 1970, 422).

One example of such an application is shown in Figure 5.16. At 100 MHz, the aromaticregion of the 1H NMR spectrum of C6H5CH2OH is a singlet, but on adding Eu(dpm)3, whichcoordinated to the OH group, the aromatic protons are shifted by amounts that depend ontheir distance from the Eu and are susceptible to first-order analysis.

Nowadays this technique is less generally used owing to the spread of high-frequencyspectrometers, but chiral reagents like a lanthanide camphorato complex find applications[Figure 5.17 shows Eu(tfc)3]. for example, when such a chiral shift reagent binds to a racemicmixture, two diasteroisomeric forms of the complex are formed, each with different peaksin the NMR spectrum; each signal can be integrated and used to calculate the enantiomericexcess, a quick way of estimating the yield of each isomer.

CF3

O

OEu

3

Figure 5.17A chiral shift reagent.

Figure 5.18 shows the 1H NMR spectrum of a mixture of 1-phenylethylamine and Yb(tfc)3

in CDCl3. The (S)-(+) and (R)-(−) isomers here give clearly distinguished peaks (the amineresonances are well downfield and not displayed).

5.5.2 Magnetic Resonance Imaging (MRI)

The use of gadolinium(III) complexes to assist diagnosis in this expanding area of medicineis now most important.

In the MRI experiment, a human body is placed on a horizontal table that is slid into thecentre of the magnetic field of the MRI scanner (essentially a pulsed FT NMR spectrometer).MRI relies on detecting the NMR signals from hydrogen atoms in water molecules (whichmake up ∼60% of the human body) and distinguishes between water molecules in healthyand diseased tissue (since water molecules in cancerous tissue have much longer relaxationtimes). MRI works well in soft tissue and has been used particularly to identify lesions inthe brain and spinal cord. When brain activity occurs, blood flow to that region increases,which causes a contrast change between the active and inactive regions.

SPH SPH


NMR Applications 79

NH2C

CH3

Ha2H3H

4H

5H

6H

15.0 14.0 13.0 12.0 11.0 10.0 9.0 8.0 7.0

Ha(R) Ha(S )

2,6(R) 2,6(S )

3,5(R) 3,5(S )

CHC13

impurity4(R)

4(S )

CH3(R)

CH3(S )

δ/ppm

Figure 5.18The 1H NMR spectrum of a mixture of 1-phenylethylamine and Yb(TFC)3 in CDCl3 (reproducedwith permission of the Editor from T. Viswanathan and A. Toland, J. Chem. Educ., 1995, 72, 945).

In order to make diagnoses about different parts of the body, it is necessary to obtain atwo-dimensional image of the signals. This is done by placing the patient in a magnetic fieldgradient (using gradient coils to create different fields at many points in a piece of tissue) sothat otherwise identical water molecules yield resonances as slightly different frequenciesat each point, depending upon their position in the field, with a computer processing thedata to give a digitized image of a spatially organized signal. The signal intensity dependsupon the relaxation times of the protons, so, in order to enhance the contrast to differentiatebetween healthy and diseased tissue, paramagnetic contrast agents are used.

5.5.3 What Makes a Good MRI Agent?

The choice is dictated by a combination of several factors:

1. High magnetic moment;2. Long electron-spin relaxation time;3. Osmolarity similar to serum;4. Low toxicity;5. High solubility in water;6. Targeting tissue;7. Coordinated water molecules;8. Large molecule with long rotational correlation times.

SPH SPH



Gadolinium(III) compounds are especially suited to this. The Gd3+ ion has a large numberof unpaired electrons (S = 7/2) and moreover its magnetic properties are isotropic. It hasa relatively long electron-spin relaxation time, ∼10−9 s, which makes it more suitable thanother very paramagnetic ions such as Dy3+, Eu3+, and Yb3+ (∼10−13 s). These factors arevery favourable for nuclear spin relaxation.

The free Gd3+(aq) ion is toxic, however, with an LD50 ∼ 0.1 mmol/kg, less than theimaging dose (normally of the order of 5 g for a human); gadolinum complexes are thereforeused, using ligands that form a very stable in vivo complex. Another factor is the presenceof water molecules in the coordination sphere of gadolinium, as relaxation times are shorterthe nearer the water molecules are to the Gd3+ ion. Ideally the more water moleculespresent the better, since more solvent waters can readily be exchanged with coordinatedwater molecules. There is a balance to be struck here, as the use of multidentate ligandsto ensure a high stability constant for the gadolinium complex, to minimize the amount oftoxic, free Gd3+ions present, tends to reduce the number of bound waters, and in practicemost contrast agents have one coordinated water molecule.

An obvious early choice for consideration as a MRI agent was the EDTA complex,[Gd(edta)(H2O)3]−. This seemed to combine several highly desirable factors, as it was sta-ble (log K = 17.35); possessed three coordinated water molecules; and utilized a cheapligand. However, it was found to be poorly tolerated in animal tests (and it was subsequentlyfound that it released significant amounts of free Gd3+, despite its high stability constant), sothe focus shifted elsewhere. The complex of the octadentate DTPA5−, [Gd(dtpa)(H2O)]2−

(gadopentetate dimeglumine; Magnevist R©; Figure 5.19) was an obvious extension(log K ∼ 22.5), and this has proved a highly successful compound, the first compoundto be approved for clinical applications (1988).

Another compound to be widely used is [Gd(dota)(H2O)]− (gadoterate meglumine;Dotarem R©; Figure 5.20), using a cyclic ligand in an attempt to form more stable com-plexes (log K ∼ 24–25). Another example is [Gd (do3a(H2O)2] (Figure 5.21).

Apart from charged species, neutral complexes are also used. One example of this is[Gd-hp-do3a] ; another is [Gd(dtpa-bma)(H2O)] (gadodiamide; Omniscan R©; Figure 5.22).In these complexes, the loss of charge is achieved by removing a carboxylate group, or by

H

Gd

H

N

N

N

O

O

O

OO

O

O

OO

O

O

Figure 5.19The structure of Gd(dtpa)(H2O)2− (reproduced by permission of Macmillan from S.A. Cotton,Lanthanides and Actinides, 1991, p. 64).

SPH SPH


NMR Applications 81

N

N C

C N

N

C

C

O

O

O

O

O

O

O

O

Gd

OH2

Figure 5.20[Gd(dota)(H2O)]−.

N

N

C

C

N

N

C

O

O

O

O

O

O

Gd

OH2

H2O

H

Figure 5.21[Gd(do3a)(H2O)2].

converting two carboxylates into amides, respectively. Both charged and neutral complexeshave advantages. Neutral complexes have less osmotic effect, so are more likely to haveosmolalities similar to blood, whilst charged species are more hydrophilic, and thereforeare more soluble.

Replacement of an NH group in [Gd-(do3a)(H2O)2] by sidechains containing OH groupswas designed to enhance solubility in the agents Gd-hp-do3a (ProHance R©; Figure 5.23)and Gd-do3a-butrol (Gadovist R©; Figure 5.24).

Another gadolinium MRI agent, of a different type, is Gadolite R© (approved for use inthe USA and UK in 1996). This is Gd3+-exchanged zeolite NaY that has been shown to bean effective contrast agent for the gastrointestinal tract, defining it from adjacent tissues.With standard MRI imaging techniques, bowel fluid in the gastrointestinal tract can mimicdiseases such as tumours. Gadolite is stable at pH 2.5–5 for several hours, and althoughgadolinium gets leached at lower pH, the oral toxicity of gadolinium is low.

5.5.4 Texaphyrins

A new type of complex with a different spectroscopic application is provided by the texa-phyrins, compounds of ‘extended’ porphyrins where the ring contains five donor nitrogens.These have attracted considerable interest because of their possible medicinal applications.

SPH SPH



N

N

N

Gd3+

-OOC

-OOC COO-

COO-

OH2

C

O

O-

N

N

N

Gd3+

-OOC

-OOC CONHCH3

CONHCH3

OH2

C

O

O-

Figure 5.22(a) [Gd(dtpa)(H2O)]2−, (b) [Gd(dtpa-bma) (H2O)].

OH

Gd

O

O

O

O

O

O

C

N

NC

CN

OH2

OH2N

Figure 5.23[Gd(hp-do3a)(H2O)].

Two texaphyrin complexes are undergoing clinical trials; a gadolinium compound (Gd-tex;XCYTRIN R©; Figure 5.25) is an effective radiation sensitizer for tumour cells. It assiststhe production of reactive oxygen-containing species, whilst the presence of the Gd3+ ionmeans that the cancerous lesions to which it localizes can be studied by MRI. It is beinginvestigated in connection with pancreatic tumours and brain cancers. One form of thelutetium analogue (LUTRIN R©) is being developed for photodynamic therapy for breastcancer, and another (ANTRIN R©) is being developed for photoangioplasty, where it haspotential for treating arteriosclerosis by removal of atherosclerotic plaque.

5.6 Electron Paramagnetic Resonance Spectroscopy

Although most lanthanide ions are paramagnetic, because of rapid relaxation effects, spectracan be obtained only at low temperatures (often 4.2 K) in most cases. From the point ofview of the chemist, EPR spectra are readily obtained (at room temperature) only fromthe f 7 Gd3+, with its 8S7/2 ground state. The sublevels of this state are degenerate in theabsence of a crystal field (in a free Gd3+ ion), but are split into four Kramers’ doublets,with MJ -values of ±1/2, ±3/2, ± 5/2 and ±7/2. The application of a magnetic field removesthe degeneracy of each doublet, and transitions can occur on irradiation with microwaveradiation, subject to the usual selection rule of �MJ = ± 1.

In the absence of a zero-field splitting (z.f.s.), all transitions occur at the same field(corresponding to a g-value of 2.00), but as the z.f.s. increases, transitions occur at higher

SPH SPH


Lanthanides as Probes in Biological Systems 83

OH

Gd

O

O

O

O

O

O

C

N

NC

CN

N

HO

OH

OH2

OH2

Figure 5.24[Gd(do3a-butrol)(H2O)].

OO

O O O

O

OMe

OMe

OH

Gd

OH

N

N

NN

N

OAcAcO

Figure 5.25Gd-tex.

and lower fields (corresponding to ‘effective’ g-values above and below 2) (Figure 5.26).The situation is not dissimilar to the high-spin Cr3+(d3) and Fe3+ and Mn2+ (d5) ions. Inthe case of the three-coordinate silylamide

[Ln[N(SiMe3]2)3

], where there is a very strong

axial distortion, signals are observed with the effective g-values g ⊥ = 8, g‖ = 2.

5.7 Lanthanides as Probes in Biological Systems

The lanthanides form a series of ions of closely related size and bonding characteristicsand in many respects resemble Ca2+, for which they often substitute isomorphously inbiological systems. Since different Ln 3+ ions can be probed with particular spectroscopictechniques (e.g. Eu3+ and Tb3+, fluorescence; Gd3+, ESR; Nd3+, electronic spectra), infavourable circumstances it should be possible to obtain information about the binding siteof spectroscopically inactive Ca2+ in several ways. Systems studied include the calcium-binding sites in calmodulin, trypsin, parvalbumin, and the Satellite tobacco necrosis virus.

SPH SPH



(a)

(b)

72

+

52

+

32

+

12

+

12

−

32

−

52

−

72

−

72

+

52

+

32

+72

±52

±

32

±12

±

12

+

12

−

32

−

52

−72

−

MJ

Increasing field

Figure 5.26Showing the effect of zero-field splitting on the EPR transitions in Gd3+.

Lanthanide chelates have been used as labels of antigens and antibodies in fluoroimmuno-logical analysis, and in areas such as the determination of steroids in animals, of hormonesand viral antigens in humans, and of herbicides in water.

Question 5.1 Work out the ground states for Pm3+ (f 4), Eu3+ (f 6), Er3+(f 11), Pr 4+(f 1) andEu2+ (f 7).Answer 5.1 Pm3+ (5I4); Eu3+ (7F0); Er3+(4I15/2); Pr4+(2F5/2); Eu2+ (8S7/2).

Question 5.2 Calculate the magnetic moment for Pr3+ ions and Dy3+ ions.

Question 5.3 Explain why f–f transitions in the electronic spectra of lanthanide complexesare weaker than d–d transitions in the corresponding spectra of transition metal complexes.

SPH SPH



Answer Forming a Ln3+ ion from a lanthanide atom by removing electrons stabilizes theorbitals in the order 4f > 5d > 6s. The 6s and 5d orbitals are emptied and the ions havethe electron configurations [Xe] 4f n . The 4f orbitals have become core-like, not involvedin bonding, with negligible crystal-field effects. Both d–d and f–f transitions are parity-forbidden. In the case of d–d spectra, vibronic coupling lowers the symmetry round themetal ion, so d/p mixing can occur and the d–d transitions are to some extent ‘allowed’.Because the 4f orbitals are so contracted compared with d orbitals, f/p mixing and vibroniccoupling are much more limited, thus the f–f transitions are less ‘allowed’ and so extinctioncoefficients of the absorption bands are smaller.

Question 5.4 Why is the Ce3+(4f 1) ion colourless whereas Ti3+ solutions (3d1) are purple?

Question 5.5 Explain why the electronic spectra of Pr3+ (4f 2) consist of a number of weaksharp lines, whereas those in the spectrum of V3+ (3d2) are stronger and broader and lessnumerous.

Question 5.6 Why do La3+, Lu3+, or Gd3+ compounds not show luminescence properties?

Question 5.7 Comment on the emission spectrum of [Eu(Me3CCOCHCOCMe3)3(2,9-dimethyl-1,10-phenanthroline)] at 77 K (Figure 5.27). It shows the 5D0 → F 7(J = 0–2)transitions. Consult Section 5.4 if required.

Comment: The 5D0 →7F0 transition consists of 2 lines (the spacing is 1.17 nm, 35 cm−1),suggesting that there are two different Eu3+ sites. There are 6 bands in the region of the5D0 →7F1 transition between 584.6 and 598.8 nm, and at least 7 bands in the region of the5D0 →7F2 transition between 610.0 and 616.6 nm.

Consulting Table 5.13 indicates that a maximum of 3 bands is predicted for the formerand 5 bands for the latter, again suggesting that there are two distinctly different europiumsites. In fact, the crystal structure of this compound shows that there are two distinctlydifferent molecules present in the unit cell, both with square antiprismatic coordination ofeuropium, with occupancy factors of 59% and 41% (see R.C. Holz and L.C. Thompson,Inorg. Chem., 1993, 32, 5251; another example of this behaviour can be found in D.F.Moser, L.C. Thompson, and V.G. Young, J. Alloys Compd., 2000, 303–304, 121.)

Question 5.8 The emission of the macrocyclic complex shown in Figure 5.28 (M = Eu)has been studied as a function of pH and metal ion concentration.A. Emission was weak over the pH range 12–8.5, but on reducing the pH below 8.5 emission

was greatly enhanced until it was almost constant between pH 6.5 and 5. On furtherreduction to pH 3, emission was gradually quenched. These changes were completelyreversible. Comment upon this.

B. The effect of adding different metals ions on the emission was investigated at pH 7.4.Groups I and II ions, as well as Zn2+, had no effect. Addition of Cu2+ ions shiftedthe absorption spectrum of the phenanthroline ligand slightly from 255 to 280 nm andcaused a substantial quenching of the europium luminescence. This was reversed, withrecovery of the europium emission, on adding EDTA solution. Comment on this.

Answer A. Excitation of the phananthroline ‘antennas’ sensitizes the Eu3+ excited state.Under alkaline conditions, this does not emit, possibly because of reduction to the

SPH SPH



570 590 610 630

λ /nm

Figure 5.27The emission spectrum of [Eu(Me3CCOCHCOCMe3)3(2,9-dimethyl-1,10-phenanthroline)] at 77 K,showing the 5D0 → FJ (J = 0–2) transitions (reproduced with permission of the American ChemicalSociety from R.C. Holz and L.C. Thompson, Inorg. Chem., 1993, 32, 5251).

N

NM

NO

N

H

O

N N

NO

N

O

NN

OH2

3+

Figure 5.28

SPH SPH



non-emitting Eu2+ state or because deprotonation of the amide group affects its abilityto populate the excited state of Eu3+. In acidic solution, basic nitrogen atoms in the phenan-throline group can be protonated, altering the electronic structure of the ligand and therebyaffecting the ability of the ligand to populate the 5D0 state of europium.Answer B. Cu2+ ions have a strong affinity for N-donor ligands like ammonia, so areexpected to bind well to phenanthroline (causing the shift in its absorption spectrum). Thephenanthroline group acts as an antenna and coordination to copper obviously affects this.EDTA, however, is a hexadentate ligand that binds copper ions more strongly, detachingthe copper from the phenanthroline group and restoring its antenna ability.Refs: T. Gunnlaugsson, J.P. Leonard, K. Senechal, and A.J. Harte, J. Am. Chem. Soc., 2003,125, 12062; T. Gunnlaugsson, J.P. Leonard, K. Senechal, and A.J. Harte, Chem. Commun.,2004, 782.

Question 5.9 List and explain the desireable qualities of a good MRI contrast agent.

Question 5.10 From your knowledge of the chemistries of calcium and the lanthanides,suggest reasons why lanthanides are good at substituting at calcium-binding sites in, e.g.,proteins.Answer 5.10 They have similar ionic radii; similar high coordination numbers; both are‘hard’ Lewis acids that prefer O-donor ligands.

SPH SPH


88

SPH SPH


6 Organometallic Chemistryof the Lanthanides


� recall that these compounds are very air- and water-sensitive;� recall that the bonding has a significant polar character;� recall that some of these compounds have small molecular structures with appropriate

volatility and solubility properties;� recall appropriate bonding modes for the organic groups;� recall that these compounds may exhibit low coordination number;� suggest suitable synthetic routes;� suggest structures for suitable examples;� recall that these compounds may be useful synthons (starting materials for synthesis);� recall that the 18-electron rule is not a predictive tool here.

6.1 Introduction

Compounds of these metals involving either σ - or π -bonds to carbon are generally muchmore reactive to both air and water than those of the d-block metals. Thus there is nolanthanide equivalent of ferrocene, an unreactive air- and heat-stable compound. They areoften thermally stable to 100 ◦C or more, but are usually decomposed immediately by air(and are not infrequently pyrophoric). Within these limitations, lanthanide organometalliccompounds have their own special features, often linked with the large size of these metals.

6.2 The +3 Oxidation State

As usual with lanthanides, this is the most important oxidation state, but there are compoundsin the 0, +2 and +4 oxidation states that are discussed later.

Simple homoleptic σ -bonded alkyls and aryls have been difficult to characterize. Theyare most readily achieved by using bulky ligands or completing the coordination spherewith neutral ligands or by forming ‘ate’ complexes.

6.2.1 Alkyls

Reaction of 3 moles of methyllithium with LnCl3 (Ln = Y, La) in THF yielded what areprobably Ln(CH3)3 contaminated with LiCl and whose IR spectra showed the presence of


SPH SPH


90 Organometallic Chemistry of the Lanthanides

L

L

L

L

M

Me

Me

Me

MeMe

Me

Li

Li

Li

L

L

Figure 6.1The structure of [Li(L−L)]3 [M(CH3)6].

THF. If an excess of CH3Li is added to MCl3 in THF, in the presence of a chelating ligand(L-L), either tetramethylethylenediamine (tmed or tmeda) or 1,2-dimethoxyethane (dme)complexes [Li(tmed)]3 [M(CH3)6] or [Li(dme)]3 [M(CH3)6] are obtained for Sc, Y, and alllanthanides except Eu (see Figure 6.1).

MCl3 + 6 MeLi +3 L−L → [Li(L−L)]3 [M(CH3)6] + 3 LiCl

A number of anionic complexes with the bulky tert-butyl ligand have been made, con-taining tetrahedral [LnBut

4]− ions {Ln = Er, Tb, Lu, confirmed (X-ray) for [Li(tmed)2]+

[LuBut4]−}.

LnCl3 + 4 But Li + n THF → [Li(THF)4][Ln(But )4] + 3 LiCl

Ln(OBut )3 + 4 But Li + 2 TMED → [Li(tmed)2][Ln(But )4] + 3 But OLi

Moving to the CH2SiMe3 ligand, the neutral alkyl [Sm(CH2SiMe3)3(THF)3] has beensynthesized and shown to have a fac-octahedral structure, whilst [Ln(CH2SiMe3)3(THF)2](Ln = Er, Yb, Lu) are five coordinate. It seems that these alkyls are extremely unstablefor lanthanides larger than Sm. Anionic species are obtained by reaction with excess ofLiCH2SiMe3 in the presence of THF or TMED to solvate the lithium ion; compounds[Li(thf)4][Ln(CH2SiMe3)4] (Ln = Y, Er, Tb, Yb) and [Li(tmed)2][Ln(CH2SiMe3)4] (Ln =Y, Er, Yb, Lu).

In order to make simple alkyls of the even bulkier −CH(SiMe3)2 ligand, an indirectapproach is needed. Direct reaction of LnCl3 with LiCH(SiMe3)2 in ether solvents givesproducts like [(Et2O)3Li(µ-Cl)Y{CH(SiMe3)2}3]; this can be avoided by using a chloride-free starting material, by using a solvent like pentane that does not coordinate to an alkalimetal, and by using a a bulky aryloxide ligand that does not bridge between a lanthanide andlithium. Thus, starting with [Ln{OC6H3But

2-2,6}3], alkyls [Ln{CH(SiMe3)2}3] have beenreported (Ln = Y, La, Pr, Nd, Sm, Er, and Lu at present). The general synthetic route is:

[Ln{OC6H3But2-2,6}3] + 3 LiCH(SiMe3)2 → [Ln{CH(SiMe3)2}3] + 3 LiOC6H3But

2-2,6

These compounds have three-coordinate triangular pyramidal structures, similar to thosefound in the silylamides [Ln{N(SiMe3)2}3].

6.2.2 Aryls

Extended reaction at room temperature between powdered Ln (Ln = Ho, Er, Tm, Lu) andPh2Hg in the presence of catalytic amounts of LnI3 affords the σ -aryls fac-LnPh3(thf)3.These have molecular structures with octahedral coordination of the lanthanides.

2 Ln + 3 Ph2Hg + 6 THF → 2 LnPh3(thf)3 + 3 Hg (Ln = Ho, Er, Tm, Lu)

SPH SPH


Cyclopentadienyls 91

[Carrying out the reaction using Eu and Yb yields EuPh2(THF)2 and YbPh2(THF)2, seeSection 6.5.1.]

Using the bulker 2,6-dimethylphenyl ligand results in [Li(THF)4] + [Ln(2,6-Me2C6H3)4]− for the two smallest lanthanides, ytterbium (lemon) and lutetium (colourless).

LuCl3 + 4 C6H3Me2Li + 4THF → [Li(THF)4][Lu(C6H3Me2)4] + 3LiCl

6.3 Cyclopentadienyls

6.3.1 Compounds of the Unsubstituted Cyclopentadienyl Ligand (C5H5 = Cp;C5Me5 = C*p)

Cyclopentadienyl (and substituted variants thereof) has been the most versatile ligand used inorganolanthanide chemistry. Unlike the situation with the d-block metals, where a maximumof two pentahapto- (η5-)cyclopentadienyls can coordinate, up to three η5-cyclopentadienylscan be found for the lanthanides, in keeping with the higher coordination numbers foundfor the f-block elements. In terms of the space occupied, a η5-cyclopentadienyl takes upthree sites in the coordination sphere. Three types of compound can be obtained, dependingupon the stoichiometry of the reaction mixture:

LnCl3 + NaCp → LnCpCl2(THF)3 + NaCl (in THF; Sm−Lu)

LnCl3 + 2 NaCp → LnCp2Cl + 2 NaCl (in THF; Gd−Lu)

LnCl3 + 3 NaCp → LnCp3 + 3 NaCl (in THF; all Ln)

The mono(cyclopentadienyl) complexes all have pseudo-octahedral structures (Figure 6.2).Another route for their synthesis is

LnCl3(THF)3 + 1/2 HgCp2 → CpLnCl2(THF)3 + 1/2 HgCl2

Cases are known of other anionic ligands present, and the triflate CpLu(O3SCF3)2(thf)3 canbe used as a synthon for a dialkyl:

CpLu(O3SCF3)2(THF)3 + 2 LiCH2SiMe3 → CpLu(CH2SiMe3)2(THF)3 + 2 LiO3SCF3

The structures of the two other families of compounds are more complicated. After synthe-sis, LnCp2Cl are isolated by removal of the THF solvent, followed by vacuum sublimation.LnCp2Cl tend to be associated, most usually as symmetric dimers [Cp2Ln(µ-Cl)2LnCp2],but DyCp2Cl has a polymeric chain structure and GdCp2Cl is a tetramer. These com-pounds can be isolated only for Gd–Lu and Y, the smaller lanthanides, for the unsubstitutedcyclopentadienyl ring, but, with the bulkier Cp* ligands, [Cp*2Ln(µ-Cl)2LnCp*2] are ob-tainable for metals like La as well. A few other systems like LnCp2X (X = Br, I) have alsobeen made.

O

OCl

ClLn

O

Figure 6.2Structure of CpLnCl2(THF)3.

SPH SPH



O ClLn

Figure 6.3Structure of [{O(CH2CH2 C5H4)2}LnCl].

Adducts [Cp*2LnCl(thf)] have been isolated from THF solutions for most lanthanidesand a number of other [Cp2LnCl(thf)] have been made. (The THF can be replaced by otherLewis bases such as MeCN and CyNC (Cy = cyclohexyl), whilst the chloride ligand maybe substituted too.) Under some circumstances, the product of synthesis is a species likethe remarkable blue, pentane-soluble ‘ate’ complex, [Cp*2Nd(µ-Cl)2Li(thf)2]. Anotherway of obtaining bis(cyclopentadienyl) complexes is to link the two rings. If the link isfunctionalized with a donor atom, coordinative saturation can be achieved without solventcoordination (Figure 6.3) and without the formation of an ‘ate’ complex, which mightotherwise be obtained (Figure 6.4).

Me2Si Li(OEt2)2

Cl

ClLn

Figure 6.4Structure of [{Me2Si(C5H4)2} In (µ-Cl)2 Li/OEt2)2].

Cases are known where these ligands bridge two metals, instead of chelating (Fig-ure 6.5).

Tris(cyclopentadienyl) compounds Ln(C5H5)3 are isolated from synthesis in THF asadducts [Ln(C5H5)3(THF)] (see below, this Section, and Figure 6.7), which undergo vacuumsublimation to afford pure Ln(C5H5)3. These compounds are thermally stable but air- andmoisture-sensitive, and are labile enough to react rapidly with FeCl2, forming ferrocene.Their coordination spheres are unsaturated, to judge by their readiness to form adducts withLewis bases like THF, RNC, RCN, and pyridine, and also by their solid-state structures.

The structures of Ln(C5H5)3 are shown in Figure 6.6. The coordination number of themetal increases with increasing size of the lanthanide. Assuming that an η5-C5H5 groupoccupies three coordination sites, then only ytterbium forms a simple molecular ‘9 co-ordinate’ [Yb(η5-C5H5)3] . The compounds of slightly larger metals like Y and Er associatethrough weak Van der Waals’ forces, affording a compound where the coordination numberis slightly greater than 9. Early lanthanides form compounds where the metals additionallyform some η1 and η2 interactions. At the other end of the series, lutetium is evidentlytoo small to form three η5-linkages. Coordination numbers of the lanthanides in thesecompounds are thus 11 for La and Pr, 10 for Nd, 9 (+) for Y, Er and Tm, 9 for Yb, and 8 for Lu.Compounds of the type Ln(C5H4R)3 can be made for substituted cyclopentadienyl ligandswhen R has little steric demand, but with bulky ligands this is frequently not possible.

Adducts [Ln(C5H5)3(THF)] exist for all lanthanides. They have ten-coordinatelanthanides with three η5-cyclopentadienyl rings and a monodentate THF, havingpseudo-tetrahedral geometry round the lanthanide (Figure 6.7). They are one of those rarecases when the same coordination number is maintained across the lanthanide series.

SPH SPH



Si

Me Me

Si

Me Me

YbBr

BrYb

Figure 6.5Structure of [{He2Si(C5H4)2}2 Yb2 Br2].

MMM

M

MM M

M

M

M

M M MM

M = La, Pr

M = Nd

M = Y, Er, Tm

M = Yb

M = Lu

Figure 6.6Structures of Ln(C5H5)3.

SPH SPH



Ln

O

Figure 6.7Structure of [Ln(C5H5)3(THF)].

Other adducts whose structure has been confirmed by X-ray diffraction include[La(C5H5)3(NCMe)2]; [Ln(C5H5)3(NCEt)] (Ln = La, Pr, Yb); [Pr(C5H5)3(CNCy)]; and[Ln(C5H5)3(py)] (Ln = Nd, Sm).

6.3.2 Compounds [LnCp*3] (Cp* = Pentamethylcyclopentadienyl)

The C5Me5 ligand is sterically more demanding than C5H5; C5Me5 also is more electron-donating than C5H5; moreover, because it is more hydrocarbon-like, C5Me5 (Cp*) com-plexes tend to be more hydrocarbon-soluble, another reason to use it for solution studies.Initially, work was concentrated on [Cp*2Sm(THF)2], which led to extremely interestingchemistry (Section 6.5.2). Attempts to synthesize [SmCp*3] and other compounds of thistype were unsuccessful, the obvious route such as the reaction between SmCl3 and NaCp*in THF leading to ring-opened products (Figure 6.8).

O

SmO

Figure 6.8Structure of the ring-opened product of the reaction between SmCl3 and NaCp∗ in THF.

In some remarakable work by W.J. Evans and his research group (W.J. Evans and B.L.Davis, Chem. Rev., 2002, 102, 2119), various syntheses have been developed which donot involve the use of THF, such as the reaction between tetramethylfulvene and a dimerichydride, and the reaction of KCp* with a cationic bis(pentamethylcyclopentadienyl) species(Figure 6.9).

The latter route has been applied to making all these compounds for La–Sm and Gd, aswell as some even more congested molecules such as [La(C5Me4R)3] (R = Et, SiMe3, Pri).The difficulty in isolating the strongly oxophilic [LnCp*3] systems is such that synthesisconditions require not only the absence of Lewis bases such as THF, RNC, and RCN, allpossibly donors, but in some cases required both the use of silylated glassware and theabsence of even traces of THF vapour from gloveboxes used.

When [SmCp*3] was eventually isolated, it was found to be [Sm(η5-C5Me5)3] (Fig-ure 6.10) and also that three of these bulky ligands fit round samarium with an increase inbond length from the usual 2.75 A to 2.82 A.

Apart from the novel ring-opening reaction with THF, [SmCp*3] inserts RNC and CO andis an ethene polymerization catalyst. Another reaction uncharacteristic of [LnCp3] systems

SPH SPH



BSm + KCp∗ Sm + KBPh4

+ SmSm

HSm

H 2 2

Figure 6.9Routes for the synthesis of [SmCp*3].

is its ready hydrogenolysis:

2 [SmCp*3] + 2 H2 → [Cp*2Sm(µ-H)2SmCp*2] + 2 Cp*H

Sm

Figure 6.10Structure of [SmCp*3].

6.3.3 Bis(cyclopentadienyl) Alkyls and Aryls LnCp2R

These compounds can be synthesized for the later lanthanides in particular. Reaction ofLnCp2Cl with LiAlMe4 gives a µ-dimethyl-bridged species (Figure 6.11). Similar, but lessstable, ethyls can also be made.

LnCp2Cl + LiAlMe4 → Cp2Ln(µ-Me2)AlMe2 + LiCl

These heterometallic dimers undergo cleavage of the bridge with pyridine (Figure 6.12) togive [LnCp2R]2.

2 Cp2Ln(µ−Me2)AlMe2 + 2 py → Cp2Ln(µ−Me2)LnCp2 + 2 AlMe3.py

Similarly, reactions of LiR with LnCp2Cl {which, strictly, should be written to give theproduct [Cp2Ln(µ-Cl)2LnCp2]} resulted in the synthesis of LnCp2R (R, e.g., Bun, But ,CH2SiMe3, p-C6H4Me, etc.) for later lanthanides:

LnCp2Cl + LiR → LnCp2R + LiCl

When this reaction is carried out in THF, a solvent that can coordinate well, the product isLnCp2R(THF).

SPH SPH



MeAl

Me

MeLn

Me

Figure 6.11Structure of [Cp2Ln(µ-Me2) AlMe2].

6.3.4 Bis(pentamethylcyclopentadienyl) Alkyls

The extra bulk of the C5Me5 ligand increases the possibility of crowding in these com-pounds. Cp*2LnR(THF) systems are less stable than the corresponding Cp2LnR(THF).[Cp*2Ln{CH(SiMe3)2}] are obtained THF-free, and even exhibit an agostic interactionbetween the lanthanide and one of the methyl carbons (Figure 6.13). The agostic Ln . . . Cdistance also decreases with increasing atomic number (size). The interaction is too weakto be seen on the NMR timescale.

Cp*2Ln(CH3) is a remarkable compound. In the solid state, it is an unsymmetricaldimer (Figure 6.14), but in solution this is in equilibrium with the monomer. MonomericCp*2Ln(CH3) undergoes some remarkable reactions with inert substances, demonstratinga very high Lewis acidity. Thus it activates CH4, a reaction that can be followed usingisotopically labelled methane.

Cp*2Ln(CH3) + 13CH4 � Cp*2Ln(13CH3) + CH4

A possible mechanism is shown in Figure 6.15. The mechanism shown above is bimolecular,involving association. An alternative, that does not involve initial coordination of methane,is shown in Figure 6.16.

Cp*2Ln(CH3) also activates benzene, and readily undergoes hydrogenolysis:

Cp*2Ln(CH3) + C6H6 → Cp*2Ln(C6H5) + CH4

2 Cp*2Ln(CH3) + 2 H2 → [Cp*2Ln(H)]2 + 2 CH4

It undergoes ready insertion of alkenes into the Ln–C bond and is a very active catalyst forthe polymerization of ethene.

Some compounds are known with just one Cp-type ligand (Figure 6.17).

[La{CH(SiMe3)2}3] + Cp*H → [Cp*La{CH(SiMe3)2}2] + CH2(SiMe3)2

It features agostic interactions with two methyl groups; if the lanthanum atom were justbound to an η5-cyclopentadienyl and also formed two an σ -bonds to alkyl groups, it wouldbe regarded as five coordinate and thus coordinatively unsaturated and electron-deficient.The agostic interactions relieve this.

Another mono(pentamethylcyclopentadienyl) alkyl is [Li(tmed)2] [Cp*LuMe3].

C

LnC

Ln

H3

H3

Figure 6.12Structure of [Cp2Ln(µ-Me2) LnCp2].

SPH SPH


CH3

CH3

CH3

H3CCH3

Si

Si

CHLn

CH3

Figure 6.13Structure of [Cp*2Ln{CH(SiMe3)2}].

2.344

2.7562.440

LuCH3Lu

H3C

Figure 6.14Structure of [Cp*2Lu(µ-Me)LuCp*2Me].

CH3Lu− CH4

+ CH4

C

H

HH

H

Lu

CH3

Lu

CH3

CH3

H δ+δ+

δ−

δ−

Figure 6.15Reaction of [Cp*2LuMe] with CH4.

CH4+LuCH2

CH

HH

CH3Lu

Figure 6.16Intramolecular elimination of methane from [Cp*2LuMe].

Si

Si

LaH3C

H3CH3C

H3C

H3C C

Si

Si

CH3

CH3

CH

CH3

CH3

CH3CH3

CH3

H H

Figure 6.17Structure of [Cp*La{CH(SiMe3)2}2].

97

SPH SPH



LuCl3+ NaCp* + 2 TMED + 3 LiMe → [Li(tmed)2] [Cp*LuMe3] + NaCl + 2 LiCl.The [Cp*LuMe3]− anion has a piano-stool structure with some asymmetry in the reportedLu–C distances of 2.39(2), 2.56(2), and 2.59(2) A.

6.3.5 Hydride Complexes

These can be made by various routes, including hydrogenolysis of Ln–C bonds; thermolysisof alkyls; and substitution of halide. Thus hydrogenolysis of the dimeric methyl-bridgecyclopentadienyl complexes (Ln, e.g., Y, Er, Yb, Lu) in THF affords [Cp2Ln(thf)(µ-H)]2:

[Cp2Ln(µ-Me)]2 + 2 H2 → Cp2Ln(THF)(µ-H)]2 + 2 CH4

A similar reaction occurs in pentane at room temperature (Ln, e.g., Y, La–Nd, Sm, Lu):

2 [Cp*2Ln{CH(SiMe3)2}] + 2 H2 → [Cp*2Ln(µ-H)]2 + CH2(SiMe3)2

The hydride group in these compounds can be very reactive:

[Cp*2LnH] + C6H6 → [Cp*2Ln(C6H5)] + H2

[Cp*2LuH] + SiMe4 → [Cp*2LnCH2SiMe3] + H2

Other hydrides are trinuclear, such as the anion [{Cp2Y(µ2-H)}3(µ3-H)]− (Figure 6.18)It is formed by the reaction of [Cp2Y(µ-Me)]2 with hydrogen in the presence of LiCl.

H

Y

Y YH

H H

−

Figure 6.18Structure of [{Cp2Y(µ2-H)}3(µ3-H)]−.

Reduction of [Cp2LuCl(THF)] with Na/THF gives a similar trinuclear species,[Na(THF)6] [{Cp2Lu(µ2-H)}3(µ3-H)].

6.4 Cyclooctatetraene Dianion Complexes

Just as C5H5− and its analogues are aromatic 6-electron ligands, the cyclooctatetraene

dianion, C8H82−, is an aromatic 10-electron ligand. Several types of complex have been

obtained. The product of the reaction between LnCl3 and K2C8H8 in THF depends uponstoichiometry:

LnCl3 + K2C8H8 → 1/2 [(C8H8)Ln(THF)Cl]2 + 2 KCl

LnCl3 + 2 K2C8H8 → K[Ln(C8H8)2] + 3 KCl

Both compounds have been obtained for most lanthanides. The anion in K(dme)2

[Ln(C8H8)2] has a ‘uranocene’-like sandwich structure (Section 13.7), whilst[(C8H8)Ce(THF)2Cl]2 has a dimeric structure with bridging chlorides (Figure 6.19). All

SPH SPH


The +2 State 99

O

Ce

Cl

O

OCe

O

Cl

Figure 6.19Structure of dimeric [(C8H8)Ce(THF)2Cl]2.

these compounds seem essentially ‘ionic’, reacting with UCl4 to form [U(C8H8)2] quanti-tatively.

Co-condensation of the metal vapour and cyclooctatraene at 77 K gives Ln2(C8H8)3

(Ln, e.g., Ce, Nd, Er); crystallization from THF gives [(C8H8)Ln(THF)2][Ln(C8H8)2].Mixed-ring complexes [(C8H8)Ln(Cp)(THF)n] can be made by routes such as equimo-lar amounts of [(C8H8)Ln(THF)Cl] and NaCp or [CpLnCl2(THF)3] with K2C8H8,though ‘one-pot’ routes work equally as well. Unsolvated complexes would be coor-dinatively unsaturated, hence the coordinated THF, as in [(C8H8)Pr(C5H5)(THF)2] and[(C8H8)Y(C5H4Me)(THF)], but use of bulky substituted cyclopentadienyl gives unsolvatedcompounds like [(C8H8)Lu{C5(CH2Ph)5}] and [(C8H8)LuCp*] (Figure 6.20).

Lu

Figure 6.20Structure of [(C8H8)LuCp*].

6.5 The +2 State

6.5.1 Alkyls and Aryls

A number of simple alkyls and aryls have been made. The use of a bulky silicon-substituted tert-butyl ligand permitted the isolation of simple monomeric (bent) alkyls[Ln{C(SiMe3)3}2] (Ln = Eu, Yb) which are easily sublimeable in vacuo. Unlike lanthanidecyclopentadienyls, it is very rare for simple σ -bonded organolanthanides to be stable enoughto be sublimeable in vacuo. Evidently the very bulky organic ligands enclose the metal ionsvery well, so that the intermolecular forces are essentially weak Van der Waals’ interactionsbetween the organic groups, ensuring the molecules are volatile at low temperatures. Othersuch compounds include [Yb{C(SiMe3)2(SiMe2X)}] (X = CH=CH2; CH2CH2OEt) andGrignard analogues [Yb{C(SiMe3)2(SiMe2X)}I.OEt2] (X = Me, CH=CH2, Ph, OMe), syn-thesized from RI and Yb. The alkylytterbium iodides have iodo-bridged dimeric structures,containing four-coordinate ytterbium when X = Me, but five coordinate for X = OMe, dueto chelation.

SPH SPH



Reaction of powdered Ln (Ln = Eu, Yb) and Ph2Hg in the presence of catalytic amountsof LnI3 affords the compounds LnPh2(THF)2. The pentafluorophenyl [Eu(C6F5)2(THF)5]has the pentagonal bipyramidal coordination geometry familiar for simple coordinationcompounds. Using a very bulky aryl ligand, Dpp (Dpp = 2,6-Ph2C6H3), results in theisolation of [Ln(Dpp)I(THF)3] and [Ln(Dpp)2(THF)2] (Ln=Eu, Yb). In [Yb(Dpp)2(THF)2]the geometry is a strongly distorted tetrahedron; there are additionally two weak η1 − π -arene interactions (Yb–C 3.138 A) involving α-carbons of the terphenyl groups. Structuresare not yet known for LnPh2(THF)2(Ln = Eu, Yb); a coordination number of four seemslow, as these are not bulky ligands. Possibly they may not have monomeric structures, orthere are agostic interactions involving Ln. . . . H–C bonding.

6.5.2 Cyclopentadienyls

Simple cyclopentadienyls [MCp2] (M = Eu, Yb) can be made by reaction of cyclopen-tadiene with solutions of the metal in liquid ammonia. The initial products are ammines[MCp2(NH3)x ], which can be desolvated by heating in vacuo. [SmCp2] is also known.These compounds have been little studied on account of their low solubilities; using insteadthe pentamethylcyclopentadienyl group confers greater solubility in aromatic hydrocarbonslike toluene.

Synthetic routes include:

SmI2(THF)2 + 2 KCp* → [Cp*2Sm(THF)2] + 2KI

EuCl3 + 3 NaCp* → [Cp*2Eu(THF)(Et2O)] + 3 NaCl + 1/2 Cp*–Cp*

These compounds often desolvate readily, thus recrystalllization of these two compoundsfrom toluene gives [Cp*2Ln(THF)] (Ln = Sm, Eu). Gentle heating of [Cp*2Sm(THF)]in vacuo forms [Cp*2Sm]. Whilst it is to be expected that [Cp*2Ln(THF)n] will have bentstructures, the angle between the planes of the Cp* rings in crystals of [Cp*2Ln] is 130–140◦.It has been suggested that a bent molecule is polarized and can give stronger dipole–dipoleforces attracting it to neighbouring molecules in the solid state. However, this is not thewhole story, as electron diffraction results indicate an angle of 160◦ in the gas phase, andanother possibility is that intramolecular nonbonded interactions are influential.

Besides forming adducts with Lewis bases (THF, Et2O, Ph3PO, etc.), [Cp*2Sm] exhibitssome remarkable reactions with dinitrogen (reversible) and carbon monoxide (irreversible)(Figure 6.21).

6.5.3 Other Compounds

Simple ‘cyclooctatetraenediyls’ [M(C8H8)] (M = Eu, Yb) can be made by reaction ofcyclooctatetraene with solutions of the metal in liquid ammonia. Another route to [M(C8H8)](Sm, Yb) is the reaction of the metals with cyclooctatetraene using iodine as catalyst. Thesecompounds are not monomers, but the Lewis base adduct [M(C8H8)(py)3] has a ‘piano-stool’ structure.

Reaction of cyclooctatetraene with solutions of ytterbium and 2 moles of potassium inliquid ammonia gives K2 [M(C8H8)2]; the solvate [K(dme)]2 [Yb(C8H8)2] (DME = 1,2-dimethoxyethane) has a sandwich uranocene-type structure for the anion.

SPH SPH


Metal–Arene Complexes 101

SmSmN

N

Sm −N2

+N2

SmCO

O

O

C

O CC

O

C

THF

THF

SmCp*2

SmCp*2

OCC

OCp*2Sm

Cp*2Sm

Figure 6.21Products of the reaction of [SmCp*2] with N2 and CO.

6.6 The +4 State

Few compounds are known in this state, confined to CeIV; even so, the oxidizing power ofCeIV, together with the tendency for species like LiR or NaR to be reducing, makes theserare. Alkoxides [Cp3Ce(OR)] (R = Pri, But ) and [Cp2Ce(OBut )2] have been synthesized.One such route is:

[Ce(OPri)4] + 3 CpSnMe3 → [Cp3Ce(OPri)] + 3 SnMe3(OPri)

This uses an organotin compound as it is a weaker reducing agent than the alkali metal com-pounds usually used, and so less likely to result in reduction to CeIII. The other compoundswith claims to be in this oxidation state are [Ce(C8H8)2] and analogues. [Ce(C8H8)2] canbe prepared by a ‘witches’ brew’ approach from a mixture of [Ce(OPri)4(PriOH)], AlEt3,and cyclooctatetraene at 140 ◦C, or by AgI oxidation of K [Ce(C8H8)2]; the latter reactioncan be reversed using metallic potassium as the reducing agent. Spectroscopic evidencesupport the view that this compound is [Ce3+{(C8H8)2}3−] rather than [Ce4+(C8H8

2−)2].The related [Ce{1,3,6-(Me3Si)3C8H5}2], more stable and more hydrocarbon-soluble, hasalso been synthesized.

6.7 Metal–Arene Complexes

A few of these have been made in which lanthanides are refluxed with arenes under Friedel–Crafts conditions, such as [(C6Me6)Sm(AlCl4)3] (Figure 6.22). These appear to have thelanthanide in ‘normal’ oxidation states; the arene group is bound η6. Interactions betweena lanthanide and an arene ring are also well-authenticated in what would otherwise becoordinatively unsaturated aryloxides, like [Yb(OC6H3Ph2)3].

A different type of compound (Figure 6.23) has been made by co-condensation of vapour-ized lanthanide atoms with 1,3,5-tri-t-butylbenzene at 77 K. These compounds are thermallystable for many lanthanides at room temperature and above, and are strongly coloured,

SPH SPH



Al ClClCl

Cl Al Cl ClClCl Sm

Cl Cl

Al

Cl Cl

Figure 6.22Structure of [(C6Me6)Sm(AlCl4)3].

But

But

But

But

But

But Ln

Figure 6.23Structure of lanthanide bis(1,3,5-tri-t-benzene) compound.

pentane-soluble solids (the yttrium and gadolinium compounds, for example, can be sub-limed at 100 ◦C/10−4 mbar with only partial decomposition).

The compounds of Nd, Gd–Er, Lu, and Y are stable to around 100 ◦C; the La, Pr, and Smcompounds decompose at 0, 40 and −30 ◦C, respectively. The compounds of Ce, Eu, Tm,and Yb could not be isolated. This cannot be explained simply by steric effects or redoxbehaviour of the metals; (i) compounds can be isolated with quite large early metals (Nd,Gd) but not with small ones like Yb; (ii) the compounds formally involve metals in the (0)state, so metals like Eu and Yb would have been expected to be successful in supportinglow oxidation states. Steric factors may have their place, as the compounds of the earlierlanthanides (La–Pr) are less stable than might be expected, possibly because the ligandsmay not be bulky enough to stabilize the lanthanide atom against decomposition pathways.

This pattern is irregular but has been correlated with the f ns2 → f n−1d1s2 promotionenergies, as it is believed that an accessible d1s2 configuration is necessary for stable bond-ing. A bonding model has been developed similar to that used for compounds like diben-zenechromium, requiring delocalization of metal e2g electrons onto the aromatic rings. Bonddissociation energies have been measured in the range of 200–300 kJ/mole, and are quitecomparable with known bis(arene) complexes of transition metals.

6.8 Carbonyls

One area which demonstrates the contrast between the lanthanides and the d-block transitionmetals is the absence of stable carbonyls. Co-condensation of the vapours of several metals(Pr, Nd, Eu, Gd, Ho, Yb) with CO in argon matrices at 4 K leads to compounds identifiedas [Ln(CO)x ] (x = 1−6) from their IR spectra. They decompose on warming above 40 K.

SPH SPH


Carbonyls 103

Other compounds containing carbonyl groups are known, involving a transition metal.Most of these feature isocarbonyl linkages. Thus [Cp*2Yb] reacts with Co2(CO)8 in THFsolution to form [Cp*2Yb(THF)-O-C-Co(CO)3]. A few compounds actually contain metal–metal bonds involving a lanthanide, thus:

Cp2LuCl + NaRuCp(CO)2 → NaCl + Cp2Lu(THF)RuCp(CO)2

The Lu–Ru bond length is 2.966 A.

Question 6.1 Suggest why [Li(L-L)]3 [Eu(CH3)6] cannot be isolated.Answer 6.1 Europium is the lanthanide with the most stable divalent state, so reductionto a EuII compound is taking place.

Question 6.2 Predict the coordination geometry of [Ln(CH2SiMe3)3(THF)2].Answer 6.2 Trigonal bipyramidal, as with main group AX5 species. This will be morestable (repulsions minimized) than a square pyramidal structure.

Question 6.3 [Yb(CH2SiMe3)3(THF)2] can be made by reaction of ytterbium chips withICH2SiMe3 in THF. Would this route be suitable for [Eu(CH2SiMe3)3(THF)2]?Answer 6.3 Because of the high stability of the EuII state, it is possible that oxidation mightonly occur as far as a europium(II) species like [Eu(CH2SiMe3)2(THF)2]. The reaction doesnot seem to have been attempted yet.

Question 6.4 Ether solutions of [Lu{CH(SiMe3)2}3] react with KX (X = Cl, Br), forming[(Et2O)K(µ-Cl)Lu{CH(SiMe3)2}3]. This reaction is not shown by the lanthanum analogue.Comment on this.Answer 6.4 Lutetium is smaller than lanthanum, so would have a greater charge densityand thus be a better electrophile, but this is still a surprising reaction. Coordination of thechloride to potassium and lutetium evidently compensates for the loss of lattice energy.

Question 6.5 Chemically [Ln{CH(SiMe3)2}3] behave as Lewis acids; besides reactingwith water, like lanthanide organometallics in general, they also are attacked by othernucleophiles such as amines and phenols, forming the silylamides and aryloxides, respec-tively. Write equations for the reactions of [Sm{CH(SiMe3)2}3] with HN(SiMe3)2 andHOC6H2But

2-2,6-Me-4.Answer 6.5 [Sm{CH(SiMe3)2}3] + 3 HOC6H2But

2-2,6-Me-4 → [Sm{OC6H2Bu t2-2,6-

Me-4}3] + 3 CH2(SiMe3)2

[Sm{CH(SiMe3)2}3] + 3 HN(SiMe3)2 → [Sm{N(SiMe3)2}3] + 3 CH2(SiMe3)2

Question 6.6 Comment on the formulae of the alkyls formed by the lanthanides with theCH3, CH2SiMe3, CMe3 and CH(SiH3)2 ligands (Section 6.2.1).Answer 6.6 The coordination number decreases from 6 in [LnMe6]3− through 5 and4 in [Ln(CH2SiMe3)3(thf)2], [LnBut

4]− and [Ln(CH2SiMe3)4]− to 3 in [Ln{(CH(SiH3)2}3],simply in order of the increasing bulk of the ligands, as hydrogens are replaced in the methylgroup by SiMe3 groups.

Question 6.7 [Cp2LnCl] are dimers in benzene solution but monomers in THF. Why?Answer 6.7 THF is a good Lewis base and cleaves the bridges, forming monomericcomplexes [Cp2LnCl(THF)]. Benzene is too weak a donor to coordinate to the lanthanide.

SPH SPH



Question 6.8 Predict the structure of the complex formed between pyrazine (1,4-diazine)and [Yb(C5H5)3].Answer 6.8 [{(C5H5)3Yb}(pyrazine){Yb(C5H5)3}] (see Figure 6.24).

NNYb Yb

Figure 6.24[(C5H5)3Yb(µ-pyrazine)Yb(C5H5)3].

LnCp2R (R e.g. Bun But , CH2SiMe3, p-C6H4Me, etc.) can be made for later lanthanides:

LnCp2Cl + LiR → LnCp2R + LiCl

When this reaction is carried out in THF, the product is LnCp2R(thf) (Figure 6.15).[Question 6.9] Structures of a number of lutetium alkyls LuCp2R(thf) have been examinedby crystallographers. Some structural data are listed in Table 6.1; comment on there.

Table 6.1 Bond lengths in some Cp2Lu complexes

Compound Lu–C (A) Lu–O (A)

Cp2Lu(But )(thf) 2.47(2) 2.31(2)Cp2Lu(CH2SiMe3)(thf) 2.376(17) 2.288(10)Cp2Lu(p-MeC6H4)(thf) 2.345(39) 2.265(28)Cp2Lu(Cl)(thf) 2.27(1)

Answer 6.9 Both the Lu–C σ -bond and Lu–O bond lengths in LuCp2But (thf) look unusu-ally long in comparison with the other compounds, suggesting that there is steric crowdinghere.

Question 6.10 The Lu–C σ -bond distance in [Cp2Lu(CH2)3NMe2] (Figure 6.25) is 2.22(1) A. Compare it with the compounds in the preceding question and comment on this.Answer 6.10 The Lu–C distance in [Cp2Lu(CH2)3NMe2] is shorter than any of the others(it is the shortest Lu–C σ -bond reported) and must reflect an uncrowded environment aroundlutetium due to the presence of a chelate ring in place of two monodentate ligands.

Question 6.11 Read Section 6.3.4 and predict the products of reaction of Cp*2Ln(CH3) ongentle warming with SiMe4, another ‘inert’ molecule.Answer 6.11 [Cp*2Ln(CH2SiMe3)3] + CH4

Question 6.12 Predict the products of reaction of [Cp*2LnH] with Et2O.Answer 6.12 [Cp*2Ln(OEt)] and EtOH.

SPH SPH


Carbonyls 105

C

Ln

N

CH2

CH2

H2

Me2

Figure 6.25Structure of [(Cp2Lu(CH2)3NMe2].

Question 6.13 Why does using the pentamethylcyclopentadienyl group make these com-pounds more soluble?Answer 6.13 The Cp* group on account of its bulk shields the metal more. It makescoordinative saturation more likely, so that there is less tendency to oligomerize, and themolecule presents a more ‘organic’ exterior to the solvent.

Question 6.14 In many of its reactions [Cp*2Sm] is oxidized to SmIII compounds. Suggestproducts of such reactions between THF solutions of [Cp*2Sm] and (i) HgPh2, (ii) Ag+

BPh4−, (iii) C5H6.

Answer 6.14 (i) [Cp*2SmPh(THF)]; (ii) [Cp*2Sm(THF)2]+ BPh4−; (iii) [Cp*2Sm(Cp)].

Question 6.15 Why do stable binary carbonyls of these metals not exist? (Consider whatd-block transition metals need to do to form stable carbonyls that the lanthanides cannotdo.)Answer 6.15 Transition metals need both vacant orbitals to accept electron pairs fromthe σ -donor (ligand) and also filled d orbitals for π back-donation (back-bonding) to theligand (note that neither the very early nor the very late transition metals form stable binarycarbonyls). CO is a rather weak σ -donor, and lanthanides tend to complex only with goodσ -donors. For π -bonding to occur, the metal must possess suitable orbitals to π -bond to theligands, and the ‘inner’ 4f orbitals are unsuited to this, unlike the dxy, dxz, and dyz orbitalsof transition metals.

SPH SPH


106

SPH SPH


7 The Misfits: Scandium, Yttrium,and Promethium


� recognize that yttrium resembles the later lanthanides in its chemistry;� recall that scandium differs from both the transition metals and the lanthanides in its

chemistry;� appreciate that promethium is a typical lanthanide in the types of compounds it forms,

despite its radioactivity and the difficulties posed to its study.

7.1 Introduction

Scandium and yttrium are elements in Group IIIA (3) of the Periodic Table, usually placedabove La (or Lu). Their treatment is frequently grouped with the lanthanides in textbooks(often for reasons of convenience). Promethium is, however, a radioactive lanthanide notencountered in the vast majority of laboratories.

Both scandium and yttrium are electropositive metals with similar reduction potentials tothe lanthanides (Eo Sc3+/Sc = −2.03 V; Eo Y3+/Y = −2.37 V; compare values of –2.37 Vand –2.30 V for La and Lu, respectively). The ionic radii of Sc3+ and Y3+ are 0.745 A and0.900 A, respectively (in six coordination). The former is much smaller than any Ln3+ ionbut yttrium is very similar to Ho3+ (radius 0.901 A); purely on size grounds, it would bepredicted that yttrium would resemble the later lanthanides but that scandium would exhibitconsiderable differences, and this expectation is largely borne out in practice. Table 7.1 listsstability constants for typical complexes of Sc3+ and Y3+, together with values for La3+

and Lu3+.

7.2 Scandium

When Mendeleev produced his original Periodic Table in 1869, he left a space for a metallicelement of atomic mass 44 preceding yttrium. The first fairly pure scandium compoundswere isolated by Cleve in 1879, but it was not until 1937 that the element itself was isolated.Although a relatively abundant element, it is fairly evenly distributed in the earth’s crustand has no important ores, though it is the main component of the rare ore thortveitite(Sc2Si2O7), thus being relatively expensive. In fact, it is mainly obtained as a by-productfrom uranium extraction.


SPH SPH


108 The Misfits: Scandium, Yttrium, and Promethium

Table 7.1 Aqueous stability constants (log β1) for complexes of Sc, Y, La and Lu

Ligand I (mol/dm3) Sc3+ Y3+ La3+ Lu3+

F− 1.0 6.2a 3.60 2.67 3.61Cl− 1.0 0 −0.1 −0.1 −0.4Br− 1.0 −0.1 −0.15 −0.2NO3

− 1.0 0.3 0.1 −0.2OH− 0.5 9.3∗ 5.4 4.7 5.8acac− 0.1 8.00 5.89 4.94 6.15EDTA4− 0.1 23.1 18.1 15.46 19.8DTPA5− 0.1 24.4 22.4 19.5 22.4OAc− 0.1 1.68 1.82 1.85

a = Data for solutions of slightly different ionic strength.

Scandium is a soft and silvery metal; it is electropositive, tarnishing rapidly in air andreacting with water. It is generally obtained by chemical reduction, for example the reactionof the trifluoride with calcium. It has only one natural isotope, 45Sc.

7.2.1 Binary Compounds of Scandium

Scandium oxide, Sc2O3, is a white solid (mp 3100 ◦C; Mn2O3 structure) made by burning themetal or by heating compounds such as the hydroxide or nitrate. It has a bcc structure with6-coordinate scandium and has amphoteric properties, dissolving in alkali as [Sc(OH)6]3−

ions. All four scandium(III) halides are known, all but the fluoride (WO3 structure) havingthe FeCl3 structure. All are white solids. They can be obtained directly from the elementsand in some cases by dehydration of the hydrated salts as well as by thermal decompositionof (NH4)3ScX6 (X = Cl, Br), a method also used for the lanthanides. Other significant binarycompounds include the fcc dihydride ScH2; unlike most of the lanthanides, scandium doesnot form a trihydride.

Scandium forms a number of well-defined lower halides, especially chlorides. Theseare synthesized by high-temperature reaction of Sc with ScCl3. They include ScCl, Sc2Cl3(‘mouse fur’), Sc5Cl8, Sc7Cl10, and Sc7Cl12. They typically have chain structures withmetal–metal bonding and octahedral clusters of scandium atoms. ScCl is isostructural withZrBr, Sc5Cl8 contains infinite chains of Sc6 octahedra sharing trans-edges, whilst Sc7Cl10

has two parallel chains of octahedra sharing a common edge. Sc7Cl12 is regarded as madeof Sc6Cl12 octahedra with isolated scandium atoms in octahedral interstices. Sc2Br3 is alsoknown.

7.3 Coordination Compounds of Scandium

As already mentioned, the ionic radius of Sc3+ is, at 0.745 A (for 6-coordination), the largestof the metals in its period, so that high coordination numbers are frequently met with in itschemistry; examples between 3 and 9 are known. 45Sc has I = 7/2 and a few NMR studiesare being reported.

7.3.1 The Aqua Ion and Hydrated Salts

For many years it was assumed that, like the succeeding d-block metals, scandium formeda [Sc(H2O)6]3+ ion. Various pieces of evidence indicated this was not the case – saltscontaining dimeric [(H2O)5Sc(OH)2Sc(H2O)5]4+ ions with approximately pentagonal

SPH SPH


Coordination Compounds of Scandium 109

Sc

H2O

H2OH2O

H2O

H2OH2O

H2OH2O

H2O

H2O OH2

OH2

OH2

OH2OH2

OH2

OH2O

OSc

H

H

4+

Sc

3+

Figure 7.1

bipyramidal seven coordination of scandium were crystallized (Figure 7.1); EXAFS spectraof solutions indicated an Sc–O distance (2.180 A) longer than that expected for octahedralcoordination (2.10–2.12 A) but in keeping with values obtained for the seven-coordinatescandium ion described above; Raman spectra of aqueous solutions were inconsistent withoctahedral [Sc(H2O)6]3+ ions; size considerations suggested seven coordination. Finally,several salts [Sc(H2O)7] X3 were obtained [X = Cl, Br, I, and C(O2SCF3)3] which contain(roughly) pentagonal bipyramidal [Sc(H2O)7]3+ ions (Figure 7.1) with Sc–O of 2.18–2.19A.

Table 7.2 gives the coordination numbers in aqueous solution of a number of [M(H2O)n]3+

ions, together with their ionic radii (for six coordination, for a fair comparison of size),showing why Sc3+ might have been predicted to form a seven-coordinate aqua ion, purelyon grounds of size.

A range of solid hydrated salts has been isolated and some structures are known.In addition to the seven-coordinate [Sc(H2O)7] X3 [X = Cl, Br, I, and C(O2SCF3)3],Sc(ClO4)3.6H2O contains [Sc(H2O)6]3+ ions and the triflate Sc(O3SCF3)3.9H2O has[Sc(H2O)9]3+ ions – as with the lanthanides (ions with these unusual coordination numbersare stabilized in the solid state by hydrogen-bonding). There are also Sc(NO3)3.4H2O andSc2(SO4)3.5H2O. The structure of the nitrate is not known, but it should be noted that eight-and nine-coordinate Sc(NO3)3(H2O)2 and Sc(NO3)3(H2O)3 molecules are present in crownether complexes.

7.3.2 Other Complexes

Scandium forms conventional halide complexes, like the hexafluoroscandates M3ScF6

which clearly contain [ScF6]3− anions, such as Na3ScF6; other octahedral anions arefound in Na3ScCl6 (cryolite structure), Cs2LiScCl6 (elpasolite structure), and Na3ScBr6.Six coordination is the norm; however, a Raman study of CsCl–ScCl3 melts in the range600–900 ◦C indicates that though the main species present is [ScCl6]3− (ν Sc–Cl 275 cm−1)there is evidence for the species [ScCl7]4− (ν Sc–Cl 260 cm−1).

A range of O-donor ligands (e.g. THF, Me2SO, R3PO) forms complexes with scandium,with two main series, ScL3X3 (where X is a coordinating anion like chloride) and ScL6X3

(with anions like perchlorate); however, at the moment, very few structures are confirmed.Six coordination is confirmed for mer-Sc(THF)3Cl3, which reacts with a small amount of

Table 7.2

M3+ ion Sc Y La Ce Pr Nd Sm Eu Gd Tb Dy

Ionic radius (A) 0.745 0.900 1.032 1.01 0.99 0.983 0.958 0.947 0.938 0.923 0.912n 7 8 9 9 9 9 9 9 8 8 8M3+ ion Ho Er Tm Yb Lu Pu Ti V Cr Fe CoIonic radius (A) 0.901 0.89 0.88 0.868 0.861 1.00 0.670 0.640 0.615 0.645 0.545n 8 8 8 8 8 9 6 6 6 6 6

SPH SPH



water to form mer-[ScCl3(THF)2(H2O)] (also octahedral), also for [Sc(Me3PO)6] (NO3)3.Likewise, scandium is six coordinate in [Sc(dmso)6](ClO4)3 and also in [Sc(dmso)6]I3,where X-ray diffraction on crystals and EXAFS studies on solutions yield similar Sc–Odistances [2.069(3) A and 2.09(1) A, respectively]. However, [Sc(pu)6](O3SCF3)3 (pu =tetrahydro-2-pyrimidone, a monodentate O-donor) has scandium in trigonal prismatic 6coordination.

There are a number of β-diketonate complexes, like Sc(acac)3; this has octahedral 6 co-ordination [like M(acac)3 for M = Ti–Co] and does not show any tendency to form adductswith Lewis bases, as do the lanthanide analogues. When nitrate is involved as a ligand,higher coordination numbers become possible, because a bidentate nitrate group takes uplittle more space in the coordination sphere than many monodentate ligands (has a small ‘biteangle’). Thus in the pentakis(nitrato)scandate ion in (NO+)2[Sc(NO3)5], nine coordinationexists with one nitrate being monodentate. Eight coordination is found in Sc(Ph3PO)2(η2-NO3)3 and [Sc(urea)4(η2-NO3)2]+ (NO3)−. Similarly, Na5[Sc(CO3)4].2H2O features fourbidentate carbonates surrounding scandium. Another way of obtaining high coordinationnumbers is to use multidentate ligands like EDTA. Crystals of NH4Sc(edta).5H2O con-tain [Sc(edta)(H2O)2]− ions, in which edta is (as usual) hexadentate and scandium is thus8 coordinate (a case where X-ray diffraction studies are essential in determining how manywater molecules are bound). Similarly, dtpa is octadentate in Mn[Sc(dtpa)].4H2O, so thatscandium is 8 coordinate here too.

Lanthanide nitrate complexes of crown ethers have been studied in detail (Sec-tion 4.3.7). Scandium nitrate does not bind directly to crown ethers; complexes likeSc(NO3)3(H2O)2.crown (crown = 15-crown-5, benzo-15-crown-5) do not involve directbonds between the oxygens in the ether ring and scandium, instead they have hydratedscandium nitrate molecules linked to the crown ether by hydrogen bonds to the ether oxy-gens. However, reaction of ScCl3 with halogen-extractor SbCl5 and 15-crown-5 in acetoni-trile gives [ScCl2(15-crown-5)](SbCl6) in which the scandium is bound to 5 ether oxygensand two chlorines in roughly pentagonal bipyramidal coordination. What has happenedis that removal of a chlorine from ScCl3 has effectively generated a linear ScCl2 moietythat is ‘threaded’ through the crown ether ring. Similar reactions occur with some othercrown ethers (benzo-15-crown-5, 18-crown-6; in the latter case the crown ether is onlypentadentate – see Figure 7.2). Carrying out the reaction in THF in the absence of crownether gives [ScCl2(THF)4]+ [SbCl5(THF)]−; in contrast, Y and La form [LnCl2(THF)5]+

[SbCl5(THF)]−.

Figure 7.2Seven-coordinate scandium in [ScCl2(18-crown-6)](SbCl6) (reproduced by permission of the RoyalSociety of Chemistry from G.R. Willey, M.T. Lakin, and N.W. Alcock, J. Chem. Soc., Chem. Commun.,1992, 1619).

SPH SPH


Coordination Compounds of Scandium 111

Sc(THF)3Cl3 is a very useful synthon (starting material) for other scandium compounds.It can conveniently be prepared using thionyl dichloride as a dehydrating agent:

ScCl3.x H2O + x SOCl2 → ScCl3(THF )3 + x SO2 + 2x HCl (reflux with THF).

Ammine complexes cannot be made by solution methods, as owing to the basicityof ammonia other species (possibly hydroxy complexes) are obtained, but a few am-mines have recently been described. (NH4)2[Sc(NH3)I5] is obtained as pink crystals fromthe reaction of NH4I and metallic scandium in a sealed tube at 500 ◦C. Sc reacts withNH4Br on heating in a sealed tantalum container, forming [Sc(NH3)2Br3], isotypic withSc(NH3)2Cl3, which has the structure [Sc2Br6(NH3)4], with isolated dimers of bromideedge-connected [Sc-mer-(NH3)3Br3] and [Sc(NH3)Br5] octahedra.[Sc(NH3)Br3], isotypicwith Sc(NH3)Cl3, is [Sc2Br6(NH3)2], with zigzag chains of edge-connected [Sc(NH3)Br5]octahedra as [Sc(NH3)1/1Br1/1Br4/2]. Scandium similarly reacts with NH4Cl in the presenceof CuCl2 to form [ScCl3(NH3)] and [ScCl3(NH3)2]. Scandium has octahedral coordinationin all of these.

7.3.3 Alkoxides and Alkylamides

Scandium resembles the 3d transition metals from titanium through cobalt in forming anunusual three-coordinate silylamide Sc[N(SiMe3)2]3 but unlike them its solid-state structureis pyramidal (Figure 7.3), not planar, in which respect it resembles the lanthanides anduranium. It does not form adducts with Lewis bases, presumably on steric grounds, theresemblance here being to the 3d metals. However, the compound of a less congested amideligand forms [Sc{N(SiHMe2)2}3(THF)], which has distorted tetrahedral four coordinationof scandium, with short Sc...........Si contacts in the solid state; this is in contrast to the5-coordinate [Ln{N(SiHMe2)2}3(THF)2].

Few scandium alkoxides have been studied in detail, but monomeric 3-coordinationexists in [(2,6-But

2-4-MeC6H2O)3Sc] where the bulky ligand enforces steric crowding. Itwill, however, expand its coordination sphere to form 4-coordinate adducts with Ph3PO andTHF – unlike the corresponding three-coordinate silylamide. Sc(OSiBut

3)3.L (L = THF,

Figure 7.3Three-coordinate Sc[N(SiMe3)2]3 (where Ln = Sc) (reproduced by permission of the Royal Societyof Chemistry from J.S. Ghotra, M.B. Hursthouse, and A.J. Welch, J. Chem. Soc., Chem. Commun.,1973, 669).

SPH SPH



Figure 7.4Conditions: (a) Me2Mg in toluene; (b) Np2Mg in toluene; (c) NaCp in THF, LiCp∗ in THF, Li(MeCp)in THF; (d) Na(Ind) in THF; (e) TMSOTf in toluene; (f) H2O in CH2Cl2; (g) LiOCMe3 in toluene; (h)LiO(Me3C6H2) in toluene; (i) NaN(TMS)2 in toluene; (j) LiCH(SiMe3)2 in toluene (reproduced withpermission from J. Arnold et al., Organometallics, 1993, 12, 3646 c© American Chemical Society2005).

py, NH3) have also been prepared (the Lewis base cannot, however, be removed to affordthe homoleptic siloxide).

Relatively few complexes of N-donor ligands are known. One of the best character-ized is with the tridentate ligand terpyridyl, which reacts with scandium nitrate to formSc(terpy)(NO3)3; here scandium is connected to 9 atoms, but one Sc–O bond is considerablylonger than the others, so a coordination number of 8.5 has been assigned. A range of por-phyrin and phthalocyanine complexes exist; syntheses often involve the high-temperatureroutes typical of the transition metals but recently a high-yield low-temperature route hasbeen utilized to make octaethylporphyrin (OEP) complexes:

Li(oep)− + ScCl3(thf)3 → Sc(oep)Cl + 2Cl− + 3 THF + Li+

The chloride can be replaced by alkoxy, alkylamide, alkyl, and cyclopentadienyl groups, asshown in Figure 7.4.

7.3.4 Patterns in Coordination Number

Table 7.3 lists in parallel corresponding binary compounds and complexes of scandium,lanthanum, and lutetium. As expected on steric grounds, the smaller scandium generally

SPH SPH


Tabl

e7.

3

Scco

mpo

und/

com

plex

La

com

poun

d/co

mpl

exL

uco

mpo

und/

com

plex

Com

poun

dFo

rmul

aC

.N.

Form

ula

C.N

.Fo

rmul

aC

.N.

Oxi

deSc

2O

36

La 2

O3

7L

u 2O

36

Fluo

ride

ScF 3

6L

aF3

9+2

LuF

39

Chl

orid

eSc

Cl 3

6L

aCl 3

9L

uCl 3

6B

rom

ide

ScB

r 36

LaB

r 39

LuB

r 36

Iodi

deSc

l 36

Lal

38

Lul

36

Ace

tyla

ceto

nate

Sc(a

cac)

36

La(

acac

) 3(H

2O

) 28

Lu(

acac

) 3(H

2O

)7

ED

TAco

mpl

exSc

(ED

TA)(

H2O

) 2−

8L

a(E

DTA

)(H

2O

) 3−

9L

u(E

DTA

)(H

2O

) 2−

8T

HF

addu

ctof

tric

hlor

ide

ScC

l 3(T

HF)

36

[LaC

l(µ

-Cl)

2(T

HF)

2] n

8L

uCl 3

(TH

F)3

6Te

rpy

com

plex

ofni

trat

eSc

(NO

3) 3

(ter

py)

8.5

La(

NO

3) 3

(ter

py)(

H2O

) 211

Lu(

NO

3) 3

(ter

py)

9A

qua

ion

[Sc(

H2O

) 7]3+

7[L

a(H

2O

) 9]3+

9[L

u(H

2O

) 8]3+

8B

is(t

rim

ethy

lsily

l)am

ide

Sc{N

(SiM

e 3) 2} 3

3L

a{N

(SiM

e 3) 2} 3

3L

u{N

(SiM

e 3) 2} 3

3Ph

3PO

com

plex

ofni

trat

eSc

(η2-N

O3) 3

(Ph 3

PO) 2

8L

a(η

1-N

O3)(η

2-N

O3) 2

(Ph 3

PO) 4

9[L

u(η

2-N

O3) 2

(Ph 3

PO) 4

]NO

38

113

SPH SPH



exhibits lower coordination numbers than the lanthanides, although sometimes the value isthe same as for lutetium, the smallest lanthanide.

7.4 Organometallic Compounds of Scandium

The organometallic chemistry of scandium is generally similar to that of the later lan-thanides. It thus forms a cyclopentadienyl ScCp3 that has mixed mono- and pentahapto-coordination like LuCp3. An anionic methyl [Li(tmed)]3[M(CH3)6] is formed by scan-dium, as by the lanthanides. However, there are often subtle differences that should beborne in mind. The pentamethylcyclopentadienyl compound [ScCp*2Me] is a monomerbut the lutetium compound is an asymmetric dimer [Cp*2Lu(µ-Me)LuCp*2Me]. Similarly,whilst triphenylscandium is obtained as a bis(thf) adduct, [ScPh3(thf)2], which has a TBPYstructure with axial thf molecules, the later lanthanides form octahedral [LnPh3(thf)3].Triphenylscandium and the phenyls of the later lanthanides are made by different routes.

ScCl3(thf)3 + 3 C6H5Li → Sc(C6H5)3(thf)2 + 3 LiCl + THF

2 Ln + 3 (C6H5)2Hg + 6 THF → 2 Ln(C6H5)3(thf)3 + 3 Hg (Ln = Ho, Er, Tm, Lu)

Thus, the synthesis of triphenylscandium is a salt-elimination reaction (or metathesis)whilst the route for the lanthanide phenyls involves a redox reaction. The former has theproblem of producing LiCl, which is often significantly soluble in organic solvents andcontaminates the desired product, whilst the latter involves disposal of mercury waste, aswell as handling toxic organomercury compounds.

7.5 Yttrium

As discussed earlier, yttrium resembles the later lanthanides strongly in its chemistry, sothat throughout this book, discussion of lanthanide chemistry includes yttrium, and itscompounds will not be examined in this chapter. To give a few, almost random, examplesof yttrium compounds strongly resembling those of the later lanthanides:

� YX3 (X = F, Cl, Br, I) have the same structures as LnX3 (Ln = Dy–Lu);� the yttrium aqua ion is [Y(H2O)8]3+ though solid yttrium triflate Y(O3SCF3)3.9H2O

contains [Y(H2O)9]3+ ions;� the acetylacetonate is [Y(acac)3(H2O)];� the bis(trimethylsilyl)amide is Y[N(SiMe3)2]3;� terpyridyl reacts with yttrium nitrate, forming 10 coordinate [Y(terpy)(NO3)3(H2O)].

Yttrium compounds are frequently useful host materials for later Ln3+ ions, as mentionedin Section 5.4.4; Eu:Y2O2S is the standard material for the red phosphor in virtually all colourand television cathode ray tubes, whilst Eu:Y2O3 is used for energy-efficient fluorescenttubes. Yttrium oxide is used to stabilize zirconia (YSZ), yttrium iron garnets (YIG) are usedin microwave devices, and of course YBa2Cu3O7 is the classic ‘warm’ superconductor.Yttrium, like scandium, is naturally monoisotopic. 89Y has I = 1/2; though signals can bedifficult to observe, valuable information can be obtained from NMR studies.

SPH SPH


Promethium 115

220.0 215.0 210.0

δ/ppm

Figure 7.589Y NMR spectrum of [Y5O(OiPr)13] in C6D6 at 25 ◦C (reproduced with permission of the AmericanChemical Society from P.S Coan, L.G. Hubert-Pjalzgraf, and H.G. Caulton, Inorg. Chem., 1992, 31,1262).

Reaction of YCl3 with lithium isopropoxide, LiOCHMe2 (LiOPri), affords an alkoxideY5O(OPri)13, which does not have a simple structure but has a cluster of 5 yttriums arrangedround a central oxygen. The 89Y NMR NMR spectrum is shown in Figure 7.5.

Two possible geometries, square pyramidal and trigonal bipyramidal, are possible for theY5O core of the molecule; the two NMR signals have relative intensities of 4:1, supportingthe square pyramidal structure. Use of bulky R groups, such as OCBut

3 or 2,6-But2C6H3,

would mean that a monomeric Y(OR)3 system is more likely.

7.6 Promethium

Mendeleev’s Periodic Table made no provision for a lanthanide series. No one could predicthow many of these elements would exist and it was not until Moseley’s work on X-ray spectrathat resulted in the concept of atomic number (1913) that it was known that an element withatomic number 61, situated between neodymium and samarium, remained to be discovered.Although several claims were made for its discovery in lanthanide ores, it was realized thatno stable isotopes of element 61 existed and, from the late 1930s, nuclear chemistry wasapplied to its synthesis.

Promethium occurs in tiny amounts in uranium ores, thus a sample of Congolese pitch-blende was found to contain (4 ± 1) × 10−15g of 147Pm per kg of ore; it was formed mainlyby spontaneous fission of 238U. It is also one of the fission products of uranium-235 and canbe obtained from a mixture of lanthanides by ion exchange. The longest-lived isotope is

SPH SPH



Pm3+(aq) Pm2(C2O4)3.xH2O(s)H2C2O4(aq)

5h

800 °C

300 °C

Pm2O3(s)

2Pm2O3(s) + 6F2(g)

500 °C

400 °C

4 PmF3(s) + 3 O2(g)

Pm2O3(s) + 6HX(g) 2 PmX3(s) + 3H2O(g) (X = Cl, Br)

PmI3(s) + 3 HX(g) (X = C1, Br)PmX3(s) + 3HI(g)

Figure 7.6Syntheses of Pm2O3, PmF3, PmCl3, PmBr3, and PmI3.

145Pm (t1/2 = 17.7 y) although a number of other isotopes exist, with both 146Pm (t1/2 = 5.5 y)

and 147Pm (t1/2 = 2.62 y) also having quite long half-lives. Because of its radioactivity andalso because it is only available in relatively small amounts, chemical studies are difficultand relatively rare. A study of the oxide and halides began with about 100 µg of the oxidePm2O3. Starting from this, all the halides were synthesized microchemically, carrying outthe reactions in quartz capillaries (Figure 7.6).

Pm3+(aq)H2C2O4(aq)−−−−−−−→ Pm2(C2O4)3.xH2O(s)

800 ◦C−−−−−−−→5 h

Pm2O3(s)

2 Pm2O3(S) + 6 F2(g)300 ◦C−−−−→ 4 PmF3(s) + 3 O2(g)

Pm2O3(s) + 6 HX(g)500 ◦C−−−−→ 2 PmX3(s) + 3 H2O(g) (X = Cl, Br)

PmX3(s) + 3 HI(g)400 ◦C−−−−→ PmI3(s) + 3 HX(g) (X = Cl, Br)

On account of the small amounts available, the compounds were characterized by non-destructive techniques – X-ray powder diffraction, Raman spectroscopy, and UV–visiblespectra. Purple-pink PmF3 (mp 1338 ◦C) has the 11-coordinate LaF3 structure, lavenderPmCl3 (mp 655 ◦C) has the 9-coordinate UCl3 structure, and coral-red PmBr3 (mp 624 ◦C)has the 8-coordinate PuBr3 structure. Red PmI3 (mp 695 ◦C) has the eight-coordinate PuBr3

structure at room temperature but the six-coordinate BiI3 structure at high temperatures.Pm2O3 itself can crystallize in one of three modifications; in the case in question, it wasobtained in the B-type structure also adopted by the Sm, Eu, and Gd oxides.

Figure 7.7 shows the UV–visible absorption spectra of three promethium halides mea-sured at room temperature. These very sharp absorption bands are characteristic of a lan-thanide; if they came from a transition metal, they would be much broader. They also showvery little dependence upon halide, showing that ligand-field effects are almost negligible;the position of the maxima would vary much more if promethium were a transition metal.(Although this is not shown in the diagram, the extinction coefficients are also much lowerthan for a transition metal.)

Other promethium compounds that have been isolated include the hydroxide Pm(OH)3

[which adopts the 9-coordinate structure of Nd(OH)3] and hydrated salts including

SPH SPH


Promethium 117

PmF3

PmCl3

PmBr3

Wavenumber × 105 m−1

Absorbance

26 24 22 20 18 16 14 12 10

Figure 7.7Solid-state absorption spectra of PmF3, PmCl3, and PmBr3 (reproduced with permission from W.K.Wilmarth et al., J. Less Common Metals, 1988, 141, 275).

PmCl3.xH2O, Pm(NO3)3.xH2O, and Pm(C2O4)3.10H2O. It would be expected that prome-thium would form some stable compounds in the +2 oxidation state, though they areunlikely to be made in aqueous solution. No definite evidence has yet been obtained, sincestudies have been hindered both by the small quantities of the element available and by itsradioactivity. The properties of promethium fit neatly into position between neodymiumand samarium; it is a microcosm of lanthanide chemistry in general.

Question 7.1 (a) Write equations for the (i) reduction of ScF3 by calcium, (ii) scan-dium oxide dissolving in potassium hydroxide solution, (iii) the thermal decompositionof (NH4)3ScCl6. (b) Why is the thermal decomposition of (NH4)3ScCl6 a good way ofmaking pure ScCl3?

SPH SPH



Answer (7.1)(a)

2 ScF3 + 3 Ca → 2 Sc + 3 CaF2

Sc2O3 + 6 OH− + 3 H2O → 2 [Sc(OH)6]3−

(NH4)3ScCl6 → ScCl3 + 3 NH4Cl

(b) No water is present that could cause hydrolysis and the other product of the reaction,ammonium chloride, sublimes on gentle heating, giving a pure product.

Question 7.2 MgCl2 melts at 712 ◦C and boils at 1712 ◦C; ScCl3 normally sublimes at800 ◦C; AlCl3 sublimes at 178 ◦C. Account for these results in terms of the covalent characterin the bonding.Answer 7.2 MgCl2 has a substantially ionic structure, giving it the highest electrostaticforces, and hence the highest melting point. ScCl3 and, even more, AlCl3 have increasingcovalent contributions to the bonding (AlCl3 sublimes as Al2Cl6 units) so that the attractiveforces between the (smaller) molecular units increase. Fajans’ rules predict that covalentcharacter increases with (1) decrease in ionic size (aluminium is smaller than scandium)and (2) increasing charge on the ion.

Question 7.3 ScH2 conducts electricity well. Suggest a reason.Answer 7.3 It is probably a scandium(III) compound Sc3+(H−)2(e−). The free electronsare in a delocalized conduction band and thus enable it to conduct electricity.

Question 7.4 Predict coordination numbers for (a) Rb2KScF6; (b) Ba2ScCl7; (c) [ScCl3{MeO(CH2)2OMe}(MeCN)]; (d) Sc(Ph2MePO)4(NO3)3; (e) ScX3(Ph3PO)4 (X = Cl, Br).Answer 7.4 (a) 6, contains [ScF6]3− ions; (b) 6, has structure Ba2[ScCl6]Cl – possi-bly not predictable; (c) 6, MeO(CH2)2OMe is bidentate and MeCN is N-bonded; (d) 8,[Sc(Ph2MePO)4(η2-NO3)2]+ (NO3)−; a monomeric Sc(Ph2MePO)4(η2-NO3)3 would be10 coordinate; (e) 6, [ScX2(Ph3PO)4] X (X = Cl, Br).

Question 7.5 Why are yttrium compounds good host materials for heavier Ln3+ ions?Answer 7.5 Because Y3+ has no f (or d electrons), its compounds have no absorptions inthe visible region of the spectrum and do not affect the magnetic properties of any addedlanthanide ions. Because its ionic radius is similar to that of Ho3+, it can accommodateseveral of the later lanthanide ions with minimal distortion.

Question 7.6 Suggest why the synthetic routes and spectroscopic methods shown for thebinary promethium compounds in Section 7.6 were employed.Answer 7.6 Only one product is a solid; by-products are gases and thus easily separatedwithout loss of the desired product (important with small quantities). The products in thecapillaries can then conveniently be used for X-ray diffraction studies (and UV–visible andRaman spectroscopy).

Question 7.7 Why are the structures of promethium compounds determined by powdermethods and not by single-crystal methods?Answer 7.7 The radiation emitted rapidly destroys the regular lattice in a single crystal.

SPH SPH


Promethium 119

Question 7.8 The visible spectrum of a promethium(III) compound (Figure 7.7) containsa considerable number of bands. How could you show that none of these were caused byimpurity?Answer 7.8 Wait for a few years (!), then remeasure the spectrum, any bands due to decayproducts will have become stronger and genuine bands due to Pm3+ will have diminished.Next, purify the promethium by ion exchange, then reprepare the compound; the impuritybands should have vanished.

Question 7.9 Explain why it is unlikely that Pm2+ compounds could be prepared by reduc-tion in aqueous solution. (Hint: consult the reduction potentials in Table 2.7). Assuming thatthis were possible, why would PmSO4 be a good choice of compound to isolate? Suggesta solvent that might be a good choice.Answer 7.9 The reduction potential of Pm2+(Pm3+/Pm2+ = −2.6 V) is such that it islikely to be a very good reducing agent and would react with water. PmSO4 is expected byanalogy with EuSO4 to be very insoluble and therefore could be followed by tracer study.A coordinating but non-aqueous solvent like THF would be a good choice.

SPH SPH


120

SPH SPH


8 The Lanthanides and Scandiumin Organic Chemistry


� recognize key lanthanide compounds used in organic reactions;� appreciate some of the reactions in which they can be used;� suggest reagents for certain transformations.

8.1 Introduction

It is a commonplace to say that there has been explosive growth in the use of lanthanidesin organic chemistry. For many years, the use of cerium(IV) compounds as oxidants waswidespread, but more recently a whole range of other compounds have made their appear-ance. Thus samarium(II) compounds are now routinely used as one-electron reducing agentsand the use of trifluoromethanesulfonate (‘triflate’) salts of scandium and the lanthanides aswater-soluble Lewis acid catalysts is widespread. Beta-diketonate complexes and alkoxideshave also come into use; there are even applications of mischmetal in organic synthesis.

8.2 Cerium(IV) Compounds

Cerium is the most stable lanthanide in the (+4) oxidation state, so CeIV compoundshave long been used as oxidants, most often as the nitrate complex, (NH4)2[Ce(NO3)6][cerium(IV) ammonium nitrate; CAN], but also (NH4)4[Ce(SO4)4] [cerium(IV) ammoniumsulphate; CAS] in particular, along with cerium trifluoracetate and others. The redox po-tential depends on the medium; in aqueous solution it is pH sensitive, getting more positive(i.e., more strongly oxidizing) the more acidic the solution. Mixtures of cerium salts withsodium bromate have been used, since the bromate is an oxidizing agent that can regen-erate the CeIV state, permitting the use of cerium in catalytic, rather than stoichiometric,quantities.

8.2.1 Oxidation of Aromatics

When working in acidic solution, arenes are generally oxidized to quinones; rea-sonable yields are obtained with symmetric hydrocarbons as starting molecules, but


SPH SPH


122 The Lanthanides and Scandium in Organic Chemistry

non-symmetrical arenes lead to several products. CAS is the reagent of choice for theseoxidations, as use of CAN may lead to introduction of nitro groups into the aromatic rings.

O

(CAS, MeCN –H2O–H2SO4; RT; 90–95%)

O

(CAS, MeCN–H2O–H2SO4; RT; 38%)

O

O

O

O

74:26

Working in other media, the rings can be functionalized instead of being converted intoquinones.

NO2

(CAN/silica gel; RT; 55%)

OH OH

NO2

OH

NO2

+

(CAN/silica gel)

OH OH OH

NO2

NO2NO2NO2

+

(CAN; CH3COOH)

CeIV Compounds oxidize side chains of aromatic compounds effectively and selectively,methylene carbons at the benzylic positions being oxidized to carbonyl groups. Polymethy-lated aromatics generally are oxidized to a single aldehyde group.

SPH SPH


Cerium(IV) Compounds 123

OCH3

CH3

CH3

OCH3

CH3

CHO

(CAN; MeOH 83%)

If alkylbenzenes are oxidized in non-aqueous media, other products are obtained; workingin ethanoic acid, acetates are formed; in alcohol, ethers result.

8.2.2 Oxidation of Alkenes

Alkenes are readily oxidized, products depending strongly upon the solvent and also uponnucleophiles present.

(CAN–I2; MeOH)

I

OCH3

(CAN–I2; ButOH)

I

ONO2

8.2.3 Oxidation of Alcohols

There are a variety of products, depending upon the alcohol. Allylic and benzylic alcoholsare easily oxidized under mild conditions. Secondary alcohols are oxidized under ratherstronger conditions. Simple primary alcohols (i.e., not ‘activated’ benzylic alcohols) are notoxidized. The oxidation of tertiary alcohols is accomplished with C–C bond fission.

C8H17C8H17

OH

OH

O

OH

(CAN/NaBrO3; H2O–MeCN; 80 °C; 88%)

PhCH2OH PhCHO (CAN; AcOH–H2O; 90 °C; 94%)

OAc

AcO AcOOH

(CAN; MeCN–H2O; 80 °C; 80%)

OAc

O

The last named is an example of a Grob oxidative fragmentation.

SPH SPH



8.3 Samarium(II) Iodide, SmI2

This is a powerful but selective 1-electron reducing agent, whose use was pioneered es-pecially by Henri B. Kagan. Soluble in solvents such as alcohols and THF, its activitycan be improved by adding a strong donor ligand, typically a Lewis base such as hexam-ethylphosphoramide [HMPA; (Me2N)3PO]. Reaction proceeds at an optimum HMPA: Smratio of 4:1, the acceleration probably involving a complex like SmI2(hmpa)4 (which hasbeen isolated). Many reactions may involve radical intermediates.

A convenient synthesis is:

Sm + ICH2CH2I → SmI2 + H2C = CH2

Carried out in THF at room temperature, this affords a deep blue solution. The effective-ness of SmI2 for some reactions can also be enhanced with certain catalysts. Thus, nickelhalides, especially NiI2, accelerate many reactions in THF, whilst iron(III) salts are used insamarium-assisted Barbier reactions. The choice of solvent is sometimes important; thuswater or alcohols are sometimes added to accelerate reactions. A disadvantage is that large,stoichiometric, amounts of SmI2 are needed, in fairly dilute mixtures, making its use costly,not least as large volumes of solvent are consumed; interest has developed in catalytic reac-tions, for example in the use of Mischmetal as a co-reducing agent for in-situ regenerationof SmI2.

8.3.1 Reduction of Halogen Compounds

Halogen compounds are readily reduced, converting RX into alkanes in good yields. Thereactivity order is I > Br > Cl. The process is very solvent-dependent; in THF only primaryhalides are reduced, but in HMPA primary, secondary, tertiary, and aryl halides are allreduced.

n-C10H21Br −→ n-C10H22 (2.5 mol SmI2 THF/HMPA, PriOH, 10 min, RT, 95%)

Sometimes coupling occurs:

PhCH2Br −→ PhCH2CH2Ph (SmI2; THF; RT; 82%)

8.3.2 Reduction of α-Heterosubstituted Ketones

This affords unsubstituted ketones

Cl

O O

(SmI2; THF/MeOH; –78 °C; 100%)

8.3.3 Reductions of Carbonyl Groups

Two outcomes are possible. In the presence of a proton source (e.g., H2O) they are reducedto alcohols, as in E.J. Corey’s approach to the synthesis of atractyligenin

SPH SPH


Samarium(II) Iodide, SmI2 125

Me

H

H

CO2Me

CO2Me

CO2Me

OTBMS

OTBMS

OTBMS

Me

H

H

HOH

Me

H

H

HOH

93

7

(SmI2; THF–H2O; RT; 97%)

O

In this synthesis, the reactant was chosen to control stereochemistry and give an equatorialalcohol, due to an intermediate axial radical; use of hydride-transfer reagents would tendto afford predominantly the axial alcohol.

Asymmetric reduction of the carbonyl group occurs in the presence of a chiral base.

(SmI2/quinidine; THF–HMPA; RT; 78%; 56% ee)PhCOCOPhPh

Ph

O

HO

In the absence of a proton source, coupling to give pinacols occurs:

C6H13C(O)CH3 C6H13

HO CH3

CH3 OH

C6H13 (SmI2; THF; RT; 80%)

This can be used to create cyclic systems containing cis-vicinal diol units.

(SmI2; THF; −78 to 0°C; 81%)

CHORO

COCH3

CO2CH3 CO2CH3CO2CH3

R = ButPh2Si

RO OH

OH

RO

OH

OH

96 : 4

Carboxylic acids and esters (and other acid derivatives) are generally inert to SmI2.However, in strongly basic or acidic conditions, reduction occurs: Carboxylic acidsand amides are reduced to primary alcohols (amides afford some amine as secondary

SPH SPH



product)

(SmI2; THF–H2O–NaOH; RT; 94%)

CH3CH2CH2CH2CH(C2H5)CO2H CH3CH2CH2CH2CH(C2H5)CH2OH

(SmI2; THF–H2O–KOH; RT; 90%)

PhCONH2 PhCH2OH + PhCH2NH2

91: 9

Nitriles are also reduced to amines

PhCN −→ PhCH2NH2 (SmI2; THF−H3PO4; RT; 99%)

8.3.4 Barbier Reactions

These involve addition of a halogenoalkane to a carbonyl compound.

BunI +

O OH

(Sml2; THF; reflux; 97%)

This reaction can be carried out rapidly at room temperature by using small amounts of Fe3+

catalyst. Considerable acceleration can be achieved by using hexamethylphosphoramide asa reaction medium (compare the following two reactions).

(2 moles SmI2; THF; reflux; 96%; 12 h)

(2 moles SmI2; THF–HMPA; RT; 92%; 1 min)

C4H9Br + C6H13COCH3 C6H13C(OH)(C4H9)CH3

C4H9Br + C6H13COCH3 C6H13C(OH)(C4H9)CH3

Intramolecular Barbier-type reactions occur between carbonyl and halogen groups in thesame molecule:

O

I

HO

(SmI2; THF; RT; Fe3+ catalyst; 90%)

(SmI2; THF; −78 to 0°C; Fe3+ catalyst; 66%)

O

I

OH

SPH SPH


Lanthanide β-Diketonates as Diels–Alder Catalysts 127

8.3.5 Reformatsky Reactions

These occur between α-bromo esters and carbonyl compounds and are often used in ring-formation reactions.

O

Br

CO2Et CO2Et+

HO

(SmI2; THF; 90%)

8.4 Lanthanide β-Diketonates as Diels–Alder Catalysts

In the Diels–Alder reaction a diene (which must be capable of being s-cis) reactswith a dienophile, which must have an electron-withdrawing group conjugated to analkene.

O

CHO

O

Lanthanide β-diketonate complexes, Ln(fod)3, widely used formerly as NMR shiftreagents (Section 5.5.1) are employed (fod = 6,6,7,7,8,8,8-heptafluoro-2,2-dimethyloctane-3,5-dionato). They have the advantages of working well under mild conditions (so that acid-sensitive groups are unaffected), as well as displaying good regio- and stereo-selectivity,with endo-adducts usually obtained.

O

+

[Yb(fod)3; RT; 24−48 h]

CHO

CHOH

H

81% 5%

+

[Yb(fod)3; toluene; 120 °C; 90%]

O

H

HO

H

H

SPH SPH



8.4.1 Hetero-Diels–Alder Reactions

Electron-rich alkenes, like the diene referred to in the first example, are very acid labile, soneed a gentle catalyst. Aldehydes are the most usual dienophiles used in these syntheses.

OSiMe3

OMe

OMe OMe

MeO

O

H+

O

O

MeO Ph

[Eu(fod)3; CH2Cl2; RT, 48h; 84%]

[Eu(fod)3; CHCl3; RT, 66%]

H3CO

+O

Me3SiOMe3SiOH

H

H

CH3CH3

CH3

CH3

Ph

8.5 Cerium(III) Chloride and Organocerium Compounds

Reaction of ‘anhydrous’ cerium(III) chloride with RLi reagents affords organocerium com-pounds {The cerium chloride is prepared by heating CeCl3(H2O)7 in vacuo up to 140 ◦Cand is in fact a monohydrate [CeCl3(H2O)], see W.J. Evans, J.D. Feldman, and J.W. Ziller,J. Am. Chem. Soc., 1996, 118, 4581.} Reaction is carried out at −78 ◦C, as decompositionis rapid at 0 ◦C, especially if a β-hydrogen is present in the R group. The exact nature ofthe cerium species is uncertain.

Some characteristics and advantages of these species are:

� They readily react with carbonyl groups by nucleophilic attack, in the same way as RLi orGrignard species, but generally react more smoothly (as they are less basic reagents thanthe alternatives, so reactions are not complicated by deprotonation side-reactions). Thesereactions afford high-yield syntheses of alcohols, especially those that can be difficult tomake by other routes.

� They do not react with other functional groups, such as halides, esters, epoxides, oramines.

� They are much less basic than RLi or Grignard reagents, and the reactions can be carriedout on molecules that tend to undergo enolization, metal-halogen exchange, reduction,or pinacol coupling when RLi is used.

� They often work in highly hindered situations, where other reagents do not avail.� Reaction with α,β-unsaturated carbonyl compounds gives 1,2-addition compounds

selectively.� They afford high diasteroselectivity, often opposite to that obtained by use of RLi. Allyl-

cerium reactants tend to react at the less substituted terminus of the allyl unit, in contrastto other allylmetallic nucleophiles, which tend to attack at the most substituted terminusof the allyl unit.

� They are cheap reactants (owing to the low cost of cerium compounds).

SPH SPH


Cerium(III) Chloride and Organocerium Compounds 129

O

O

Ph Ph Ph Ph

HO Bun

But

(CeCl3, BunMgBr; THF; 0°C; 98%)

OMe

OMe

MeO

MeO HO

(ButLi/CeCl3; THF; 42%)

Li /CeCl3

MeCH=CHCHO+

HO

82%

−

OOH

Pri

(CeCl3, PriMgCl; 72%, 3% yield if just Grignard)

(CeCl3, PhMgBr; THF–H2O; RT; 97%)

Ph CH CH C Ph

Ph CH CH2 C Ph

O

Ph

Ph CH CH C

O

Ph

H

89% (5% if Grignard alone)

11% (81% if Grignard alone)

Ph

O

Reactions of chiral cycloalkenyl reagents with rigid ketones affords products with a highdegree of stereodifferentiation:

O

+

Li/CeCl3

OH

−78°C

THF86%

OH

12 : 1

+

SPH SPH



8.6 Cerium(III) Chloride and Metal Hydrides

In combination with reducing agents (NaBH4, LiAlH4), cerium chloride modifies the re-ducing ability of the hydride. In combination with NaBH4, CeCl3.7H2O selectively reducesthe C=O group in enones, when use of NaBH4 by itself would give a mixture of allylicalcohols and saturated alcohols (the Luche reaction).

O OHH

(CeCl3–NaBH4; MeOH; 0°C; 88%)

This mixture can be used for the stereoselective reduction of saturated ketones

(CeCl3– NaBH4; EtOH; −78°C; ∼100%)

OPh

PO

Ph Ph

Ph

PhPh

OP

OH Ph

PhPh

OP

OH

+

>95% <5%

and for the selective reduction of ketones in the presence of aldehydes, since, in alcoholicsolution, the Ce3+ ion catalyses acetalization of aldehydes, protecting them from reduction.

(CeCl3– NaBH4; EtOH/H2O; 75%)

O

CHO CHO

CO2Me CO2MeHO

A mixture of LiAlH4 and CeCl3 is a powerful reducing agent, reducing unsaturatedcarbonyl compounds to allylic alcohols (in this case, anhydrous cerium chloride is a ne-cessity). It will reduce phosphine oxides to phosphines, oximes to primary amines, andα,β-unsaturated carbonyl compounds to allylic alcohols. In particular, it reduces both aryland alkyl halides to hydrocarbons.

(LiAlH4–CeCl3; DME; heat, hn; 94%)

F

P

O

Me

Ph

P Me

Ph

(LiAlH4–CeCl3; THF; 40°C; 89%)

SPH SPH


Scandium Triflate and Lanthanide Triflates 131

8.7 Scandium Triflate and Lanthanide Triflates

These reagents are associated especially with S. Kobayashi. Ln(O3SCF3)3 [Ln(OTf)3] aresynthesized in aqueous solution, crystallized as nonahydrates (Section 4.3.2) and the an-hydrous triflates can be prepared by heating the hydrates. They are very effective Lewisacid catalysts, which perform well under mild conditions and can be used in both aque-ous and organic solutions. Whilst most Lewis acids are decomposed in water (strictly dryorganic solvents are employed in these reactions, cf. AlCl3 in Friedel–Crafts reactions), tri-flates are not affected. The ability to avoid the use of organic solvents gives triflates powerful‘green’ credentials. They are catalysts for many reactions (Michael, Diels–Alder, allylation,Friedel–Crafts, imino-Diels–Alder, etc.). Scandium triflate has attracted especial attention,though the use of the cheaper ytterbium compound has also been well investigated.

8.7.1 Friedel–Crafts Reactions

Small amounts of Sc(OTf )3 [as well as many Ln(OTf )3] catalyse some Friedel–Crafts re-actions. Unlike conventional Friedel–Crafts catalysts, there is no need to use stoichiometricamounts. The yield does depend upon the lanthanide; for example, in the following reaction,89% (Sc), 55% (Yb), 28% (Y).

MeO + (CH3CO)2O MeO COCH3

[M(OTf )3 cat; CH3NO2; 50°C]

Yb(OTf)3 is an excellent catalyst for the aldol reactions of silylenol ethers with aldehydesin aqueous solution, working better than in organic solvents like THF and MeCN, thoughthe reactions can also be performed in organic solvents, and, after the reaction has beenquenched by the addition of water, the triflate catalyst may be recovered from the aqueouslayer.

Ph

Me3SiO+ HCHO

Ph

O

OH

[Yb(OTf )3 cat; H2O–THF; 94%]

Even greater rates have been achieved in the Yb(OTf)3-catalysed aldol reactions ofsilylenol ethers with aldehydes in micelles, by adding a small quantity of surfactant, suchas sodium dodecyl sulfate.

A study of the reaction of benzaldehyde with 1-(trimethylsilyloxy)cyclohexane showsa remarkable dependence of the yield upon the lanthanide used, being as high as 91% forYb(OTf )3 and as low as 8% with La(OTf )3.

+ PhCHO

[M(OTf )3 cat; H2O–THF; RT]

OSiMe3 OOH

Ph

SPH SPH



8.7.2 Diels–Alder Reactions

Lanthanide triflates catalyse Diels–Alder reactions, with the scandium complex as the mosteffective catalyst, and, again, the catalyst can be recovered and reused, being just as effectivein subsequent runs.

endo/exo = 100/0

[Sc(OTf )3 cat.; THF–H2O; RT; 93%]

O

O

H

H

+

O

O

+

[Sc(OTf )3 cat.; THF–H2O; RT; 83%]

endo/exo = 95/5O

O

The triflates catalyse the allylation of carbonyl compounds with tetraallyltin, producingintermediates in the synthesis of higher sugars.

+PhCHO Sn

4 Ph

OH

[Yb(OTf )3 cat.; H2O–EtOH–toluene; RT; 90%]

8.7.3 Mannich Reactions

Another synthesis to which they have been applied is Mannich-type reactions betweenimines and enolates (especially silyl enolates) to afford β-amino ketones or esters.

+

[Yb(OTf)3 (cat.); CH2Cl2; 0°C; 97%]

N

HPh

Ph OSiMe3

OMe

NH O

Me

Ph

Ph

Since many imines are not stable enough for their isolation and purification, an extensionof the previous synthesis lies in a one-pot reaction that reacts with in situ prepared imines

SPH SPH


Scandium Triflate and Lanthanide Triflates 133

with silyl enolates to produce β-amino esters.

NH O

OMe

Bu

Ph

PhCHO + BuNH2 +OSiMe3

OMe

[Yb(OTf )3 (cat.) + dehydrating agent; CH2Cl2; RT; 85%]

β-Amino ketones can be made from an aldehyde, an amine, and a vinyl ether in a one-potsynthesis, an aqueous Mannich-type reaction.

NH Op -ClC6H4

Ph

PhCHO + p-ClC6H4NH2 +

[Yb(OTf )3 (cat.); THF–H2O; RT; 90%]

Me

OMeMe

8.7.4 Imino-Diels–Alder Reactions

Lanthanide triflates are catalysts for a powerful imino-Diels–Alder reaction for buildingN-containing six-membered heterocycles.

[Yb(OTf )3 (cat.); CH3CN; RT; 69%]

[Yb(OTf )3 (cat.); CH3CN; RT; 95%]

N

HPh

+NH

H

H

Ph

N

H

Cl

Ph

+NH

PhCl

OEt

OEt

H

The yields again depend upon the choice of the lanthanide triflate. In a study of thereaction:

[M(OTf )3 (cat.); CH3CN; RT]

N

HPh

+NH

Ph

H

H

best yields were obtained with heavy lanthanides such as Er (97%) and Yb (85%), in contrastto La (45%) and Pr (60%).

SPH SPH



Another one-pot synthesis uses reaction of an aldehyde, an amine, and an alkene to givepyridine and quinoline derivatives via the imino-Diels–Alder route.

SPh++

NH

CO2Me

Cl

SPh

MeO2CCHO

[Yb(OTf )3(cat.); CH3CN; RT; 65%]

NH2

Cl

Extensions of the triflate catalyst include the use of versions with fluorinated ‘ponytails’such as [Sc{C(SO2C8F17)3}3] for use in fluorous phase Diels–Alder reactions (fluorinatedsolvent.)

8.8 Alkoxides and Aryloxides

Complexes LnM3tris(binapthoxide) (M = alkali metal) have a good deal of utility ascatalysts.

Li

(thf )2Li

O

O

O

O

O O

Ln

Li(thf )2

(thf )2

The lanthanide complexes are soluble in ethers like Et2O or THF. They are catalystsfor the enantioselective reduction of aldehydes, with best results for the lanthanum com-plex. Thus the yield is 72% when M = La, but 50% when M = Yb; more strikingly, theenantiomeric excess is 69% for the La complex but only 3% when the Yb complex was

SPH SPH


Lanthanide Metals 135

used.

(MeLi/Li3La(S-binol)3; Et2O; °C) -98

H

O

C6H5 CH3C6H5

OH

69% ee

The research group of Shibasaki has studied similar compounds, LnM3(binapthoxide)3.x THF.yH2O (M = Li, Na, K) as catalysts for a wide range of reactions, including epox-idation of enones, hydrophosphonylation of imines and aldehydes, and a range of asym-metric C−C bond-forming reactions (Diels–Alder, Michael addition, aldol and nitroal-dol formation). Yields and enantioselectivity can depend markedly upon the alkali metal.Thus, in the nitroaldol reaction shown below, the yields are very similar (90, 92%) whenLnM3(binapthoxide)3 are used (M = Li, Na), however the ee is 94% when using the Licompound but only 2% if the sodium compound is employed.

O

CHO

MeO

+ MeNO2 O

MeOHO

NO2

Lanthanide isopropoxides, usually written Ln(OPri)3, but more likely to be oxo-centredclusters Ln5O(OPri)13, are used, not just as starting materials for the synthesis of catalystssuch as the naphthoxides but also as catalysts in their own right. They have been used in theMeerwein–Ponndorf–Verley reaction, where carbonyl compounds are reduced to alcohols,recent studies having shown that the reaction takes place exclusively by a carbon-to-carbonhydrogen transfer.

8.9 Lanthanide Metals

Apart from the role of mischmetal as a co-reductant in SmI2-catalysed reactions such aspinacolization (Section 8.3.3) mischmetal has been found to be a useful reactant in its ownright. Thus a combination of mischmetal and additives (1,2-diiodoethane or iodine) has beenactive in Reformatsky-type reactions. Though giving slightly lower yields than SmI2, theyare less air-sensitive; there is the further handicap of requiring slow (dropwise) addition ofreactants.

O +Br

Et

CO2EtOH

Et

CO2Et

(mischmetal + additive; THF; RT; then acidify)

Lanthanum–nickel alloys are useful hydrogenation catalysts. The alloy LaNi5 readilyabsorbs large amounts of hydrogen. The catalysts are robust, hard to poison, are capable ofrepeated reuse, and operate under mild conditions.

SPH SPH



O

O

O O

O

O

(LaNi5H6; THF–MeOH; RT; 93%)

(LaNi5H6; THF–MeOH; RT; 78%)

CO2Me CO2Me

It selectively catalyses hydrogenation of C=C units (especially conjugated double bonds)but under more forcing conditions it also reduces other functional groups such as carbonyl,nitro, and nitrile.

(LaNi5H6; THF−MeOH; 0 °C; 6 h; 89%)

O O

Compare

CHO

(LaNi5H6; THF−MeOH; 0 °C; 3 h; 89%)

CHO

and

(LaNi5H6; THF−MeOH; RT; 12 h; 95%)

CHO CH2OH

8.10 Organometallics and Catalysis

Considerable study is being made of the ability of organometallic compounds of the lan-thanides (and actinides) to catalyse reactions of organic substrates, much of this workassociated with the name of Tobin J. Marks. The hydride [Cp*2Lu(µ-H)2LuCp*2] (Cp* =C5Me5) is an exceptionally active catalyst for the homogeneous hydrogenation of alkenesand alkynes. A suggested mechanism, involving dissociation to the coordinatively unsatu-rated monomer [Cp*2LuH], is shown below (Figure 8.1) – readers will note a resemblanceto the mechanism of hydrogenation using [RhCl(PPh3)3].

SPH SPH


Organometallics and Catalysis 137

A whole range of reactions are catalysed by such complexes. Figure 8.2 shows the mech-anism proposed in a recent study of hydroamination, catalysed by [Cp∗

2La{CH(SiMe3)2}2].It indicates a two-stage mechanism, cyclization to form La–C and C–N bonds, followedby La–C protonolysis. The alkene inserts into the La–C bond via a four-centre transitionstate, followed by protonolysis by a second substrate molecule and dissociation of the cy-clized amine, which regenerates the catalyst. The rate-determining step involves a highlyorganised seven-membered chair-like cyclic transition state.

Lu H

LuH

H

Lu

LuH

R

Lu

R

HH

Lu

R

H2

R

Figure 8.1Catalytic hydrogenation using [Cp*2LuH] catalyst.

SPH SPH



H2N n

CH2(SiMe3)2

La N

H

n

La N

H

n

N

H

Lan

N

H

n

H2N n

La CH(SiMe3)2

Figure 8.2Hydroamination catalysed by [Cp*2La(CH(SiMe3)2)].

Questions8.1 How is SmI2 prepared? How may its activity be enhanced? What are the advantages

to using it? What are the drawbacks to using it, and how can these be circumvented?

8.2 What are the main uses of lanthanide triflates? What are the advantages to usingthem?

8.3A How are organocerium compounds prepared? Why are they useful? Why canCeCl3/NaBH4 be used in the presence of water, whilst CeCl3/LiAlH4 requires strictlyanhydrous conditions?

8.3B The complexes LnM3tris(binapthoxide) (M = alkali metal) have a good deal of utilityas catalysts (Section 8.3). They are readily prepared from either the appropriatelanthanide silylamide, Ln[N(SiMe3)2]3, or chloride. Why choose the former, whenthe chloride can be bought ‘off the shelf’.Answer : The former method has the advantage of a cleaner reaction, a lanthanidestarting material soluble in common organic solvents, and chloride-free products.

SPH SPH



Problems8.4 Suggest a product for:

CH3

OOAc

OAcAcO

CH3

O

O

(CAN; MeCN–H2O)

Answer:

CHO

OOAc

OAcAcO

CH3

O

O

8.5 Suggest a product for:

OH

OH

[CAN (10%)–NaBrO3; MeCN–H2O; 80°C]

Answer:

OH

O

8.6 Suggest a product for this reaction:

CH2CH3

(CAN; 3.5M HNO3; 70°C)

Answer:

O

CH3

[G.A. Molander, Chem. Rev., 1992, 92, 29 (in particular, p. 31)].

SPH SPH



8.7 Suggest a product of this reaction:

OH

OH

(CAN; AcOH–H2O; 25°C)

Answer:

O(2 molecules)

(T. Imamoto, Lanthanides in Organic Synthesis, Academic Press, 1994, p. 130; G.A.Molander, Chem. Rev. 1992, 92, 29).

8.8 Explain the difference between these reactions:

HOO

OOH

(CAN; MeCN–H2O; 60°C; 90%)

(CAN; MeCN–H2O; 60°C; 85%)

Answer: In one isomer, there is the possibility of delta-hydrogen abstraction by analkoxyl radical, forming a furan ring; in the other, the alcohol group is too far awayfrom the delta hydrogen for this to happen.

8.9 Predict the two products of:

O

+ H2C •H

CO2Et[Eu(fod)3; RT; 20 h]

Answer:

O

CO2Et

H

H

H O

H

CO2Et

H

H

75% yield

89 : 11

[G.A. Molander, Chem. Rev., 1992, 92, 29 (in particular, p. 60)].

SPH SPH



8.10 Predict the product of this reaction:

OH

CH3

EtO

+ [Yb (fod)3; neat; RT, 60–80%]

Answer:

O

H CH3

OEt

H

(T. Imamoto 107, Lanthanides in Orgranic Synthesis, Academic Press, 1994, 107).


O(CeCl3, PriMgBr; THF; 0°C)

Answer: (Pri)3COHG.A. Molander, Chem. Rev., 1992, 92, 29 [(in particular, p. 44)].


(LiAlH4–CeCl3; THF; heat)CH3(CH2)11F

Answer: C12H26

[G.A Molander, Chem. Rev., 1992, 92, 29 (in particular, p. 37)].


(LiAlH4–CeCl3; THF; RT)NOH

PhPh

SPH SPH



Answer:

NH2

PhPh

[G.A Molander, Chem. Rev., 1992, 92, 29 (in particular, p. 39)].


+ [Sc(OTf )3 (cat.); THF–H2O; RT]

O

Answer:

O

[S. Kobayashi (ed.) Lanthanides: Chemistry and use in Organic Synthesis, Springer-Verlag, Heidelberg, 1999, p. 78].


+

[Sc(OTf )3 (cat.); THF–H2O;RT]

N O

OO

Answer:

O

OCON

[S. Kobayashi (ed.) Lanthanides: Chemistry and Use in Organic Synthesis, Springer-Verlag, Heidelberg, 1999, p. 81].

SPH SPH




+ Sn4

[Sc(OTf )3 (cat.); H2O–THF; RT]N

OH

CHO

Answer:

N

OH

OH



+ [Yb(OTf )3 (cat.); CH3CN; RT]+PhCOCHO

NH2

OMe

Answer:

NH

MeO

COPh

H

H


SPH SPH


144

...


9 Introduction to the Actinides


� recall that most of the actinides do not occur in nature, that all are radioactive, and thatthey have increasingly short half-lives as the series is traversed;

� understand the methods used to extract and synthesize them;� understand isotopic separation processes and the need for them;� recall that the early actinides exhibit transition-metal-like behaviour and that the later

actinides resemble the lanthanides in their chemistry;� explain the stability of the different oxidation states for the actinides;� recognize +4 and +6 as the main oxidation states for uranium, and +4 as the only

important state for thorium;� recall that relativistic effects are significant in the chemistry of these elements.

9.1 Introduction and Occurrence of the Actinides

Although the first compounds of uranium and thorium were discovered in 1789 (Klaproth)/1828 (Berzelius), most of these elements are man-made products of the 20th century.Thorium and uranium are both long-lived and present in the earth in significant amounts,but for practical purposes the others should be regarded as man-made, though actiniumand protactinium occur in nature in extremely small amounts, as decay products of 235Uand 238U, whilst microscopic amounts of plutonium, generated through neutron capture byuranium, have been reported to occur naturally. The principal thorium ore is monazite, aphosphate ore also containing large amounts of the lanthanides, whilst the main uraniumore is U3O8, usually known as pitchblende. Elements beyond uranium are man-made andtheir synthesis is described in section 9.2. All the actinides are radioactive, sometimesintensely; this requires special care in their handling, and the radioactivity often plays apart in their chemistry, as in causing radiation damage in solutions and in dislocating theregular arrangements of particles in crystals.

9.2 Synthesis

Apart from thorium, protactinium, and uranium, the actinides are obtained by a bombard-ment process of some kind. Thus for actinium (and also protactinium):

226Ra + 1n → 227Ra + γ ; 227Ra → 227Ac + β−

230Th + 1n → 231Th + γ ; 231Th → 231Pa + β−


...



Similarly, for the elements directly beyond uranium:

238U + 1n → 239U + γ ; 239U → 239Np + β−; 239Np → 239Pu + β−

and 238U + 1n → 237U + 21n; 237U → 237Np + β−

For elements beyond plutonium, successive neutron capture is required. This is, obviously,a slower process; thus starting with 1 kg of 239Pu, using a neutron flux of 3 × 1014 neutronscm−2s−1 (a high but experimentally feasible value), around 1 mg of 252Cf is obtained after5–10 y.

239Pun−→ 240Pu

n−→ 241Pun−→ 242Pu

n−→ 243Pun,β−−→ 243Am

n−→ 244Am

↓ β− ↓ β−

241Amn,γ−→ 242Am

β−−→ 242Cm 244Cm

242Pu or 243Am or 244Cmmultiple n, β−

−−−−−−→ 249Bk, 252Cf, 253Es, 257Fm

The limit of this process in a reactor is 257Fm; the product of the next neutron absorption is258Fm, which undergoes spontaneous fission (t1/2 = 0.38 ms). This point can be passed intwo ways. One is to utilize a more intense neutron flux than can be obtained in a reactor, inthe form of a thermonuclear explosion, so that a product such as 258Fm undergoes furtherneutron absorption before fission can occur. Here, in the synthesis of 255Fm in ’Ivy Mike’, theworld’s first thermonuclear test, at Eniwetok atoll on 1st November 1952, the initial productof multiple neutron capture, 255U, underwent a whole series of rapid decays, yielding 255Fm.

238U17 n−→ 255U

β−−→ 255Np

β−−→ 255Pu

β−−→ 255Am

β−−→ 255Cm

β−−→ 255Bk

↓ β−

255Fmβ−

←− 255Esβ−

←− 255Cf

For obvious reasons, this route is not likely to be followed in the future; instead, heavy-ion-bombardment, using particles such as 11B, 12C, and 16O, will be used, though morerecently heavier ones like 48Ca and 56Fe have been utilized. This route is reliable but hasthe twin drawbacks (from the point of view of yield) of requiring a suitable actinide targetand also being an atom-at-a-time route. Examples include:

244Cm + 4He → 247Bk + 1H

253Es + 4He → 256Md + 1n

248Cm + 18O → 259No + 3 1n + 4He249Cf + 12C → 255No + 2 1n + 4He

248Cm + 15N → 260Lr + 3 1n

249Cf + 11B → 256Lr + 4 1n

The most important and longest-lived isotopes are listed in Table 9.1.

...


Extraction of Th, Pa, and U 147

Table 9.1 Longest-lived isotopes of the actinide elements

Mass Half-life Means of decay

Ac 227 21.77 y β−Th 232 1.41 × 1010 y α

Pa 231 3.28 × 104 y α

U 234 2.45 × 105 y α

235 7.04 × 108 y α

238 4.47 × 109 y α

Np 236 1.55 × 105 y β−, EC237 2.14 × 106 y α

Pu 239 2.14 × 104 y α

240 6.57 × 103 y α

242 3.76 × 105 y α

244 8.26 × 107 y αAm 241 432.7 y α

243 7.38 × 103 y αCm 244 18.11 y α

245 8.5 × 103 y α

246 4.73 × 103 y α

247 1.56 × 107 y α

248 3.4 × 105 y α

250 1.13 × 104 y α, SFBk 247 1.38 × 103 y α

249 320 d β−Cf 249 351 y α

250 13.1 y α251 898 y α252 2.64 y α

Es 252 472 d α253 20.47 d α254 276 d α255 39.8 d β−

Fm 257 100.4 d αMd 258 55 d αNo 259 1 h α, ECLr 260 3.0 min α

9.3 Extraction of Th, Pa, and U

9.3.1 Extraction of Thorium

Thorium makes up around 10% of monazite; one process for its extraction relies on treatingthe ground ore with concentrated aq. NaOH at 240 ◦C for some hours, then adding hotwater. This dissolves out Na3PO4, leaving a mixture of the hydrated oxides and hydroxidesof the lanthanide and Th. On adjusting the pH to 3.5 with boiling HCl, the lanthanide oxidesdissolve, leaving behind only hydrated ThO2. This can be converted into Th(NO3)4, andpurified by solvent extraction into kerosene with tributyl phosphate.

In the acid route, the phosphate ore is digested with sulfuric acid at 230 ◦C for somehours and thorium phosphate is precipitated on adjusting the pH to 1. This is purified byfirst converting it into Th(OH)4, thence into thorium nitrate, purified as above.

...



9.3.2 Extraction of Protactinium

First identified in 1913 (the first compound, Pa2O5, was isolated in 1927 by von Grosse,who isolated the element in 1931), protactinium is not generally extracted. Most of what isknown about the chemistry of protactinium ultimately results from the extraction in 1960by the UK Atomic Energy Authority of some 125 grams of Pa, from 60 tons of wastematerial left over from the extraction of uranium, at a cost of about $500,000 (very roughly,£1,250,000 at today’s exchange rate).

9.3.3 Extraction and Purification of Uranium

Uranium is deposited widely in the Earth’s crust, hence it has few ores, notably the oxidesuraninite and pitchblende. The ores are leached with H2SO4 in the presence of an oxidizingagent such as NaClO3 or MnO2, to oxidize all the uranium to the (+6) state as a sulfate orchloride complex. On neutralization with ammonia a precipitate of ‘yellow cake’, a yellowsolid with the approximate composition (NH4)2U2O7 is formed. This is converted into UO3

on ignition at 300 ◦C. This can be purified further by conversion into uranyl nitrate, followedby solvent extraction using tributyl phosphate in kerosene as the extractant.

9.4 Uranium Isotope Separation

Having purified the uranium, it is then treated to separate the 235U and 238U isotopes for nu-clear fuel purposes (any uranium compounds purchased commercially are already depletedin 235U). In practice, nuclear fuel requires enrichment from the natural abundance of 0.71 %235U to around 5%, so what follows details a degree of enrichment not usually required.

The uranium compound usually used is UF6. It is chosen on account of its volatility(sublimes at 56.5 ◦C) and low molecular mass (Mr), despite its extreme sensitivity tomoisture (and toxicity of the HF produced) requiring the use of scrupulously sealed andwater-free conditions, as well as fluorine-resistant materials.

9.4.1 Gaseous Diffusion

UF6 vapour diffuses through barriers (F2-resistant Al or Ni) with pores of diameter ca 10–25nm at 70–80 ◦C.

Applying Graham’s Law dictates a separation factor, α*, where

α∗ = √Mr(

238UF6)/Mr(235UF6) = √

352/349 = 1.00429

Consecutive stages are linked in a cascade to provide the desired degree of enrichment, with3000 stages giving up to 90% enrichment in 235U. A great deal of energy is needed to pumpthe UF6 round the system.

9.4.2 Gas Centrifuge

On centrifuging UF6 vapour in a gas centrifuge, 238UF6 tends to concentrate towards theoutside of the centrifuge whilst the lighter 235UF6 predominates near the axis of the cen-trifuge. Better separations are achieved at high speeds and temperatures, and an advantageof the process is that the separation is in direct proportion to Mr.

...


Characteristics of the Actinides 149

9.4.3 Electromagnetic Separation

Ionized UCl4 was separated using a cyclotron-like system to obtain the first enriched samplesused in the Manhattan project. In practice, this was a difficult method to utilize and wasrejected in favour of gaseous diffusion.

9.4.4 Laser Separation

One process has involved selectively ionizing 235U in uranium vapour using a tuneable laser.In another process (Figure 10.6), 235UF6 molecules in a gaseous mixture are selectivelyionized using an IR laser (the isotope shift of ν U-F is around 0.5 cm−1); the excited 235UF6

molecules are decomposed by a UV laser to give solid UF5; the 238UF6 molecules areunaffected. Low temperatures are used to give sharp absorption bands so that the vibrationof the 235UF6 molecules but not that of the 238UF6 molecules is activated. Decompositionof U(OMe)6 has also been studied.

9.5 Characteristics of the Actinides

Table 9.2 depicts the oxidation states of the actinides. For the early actinides, Ac–Np, thehighest (though not necessarily the most stable) oxidation state reflects the total numberof electrons (6d and 5f) that can be removed from the outer shell. This resemblance tothe transition metals was noted for Ac–U nearly a century ago and initially made peoplethink that the actinides were another block of transition metals. Later, starting round aboutBk, most elements tend to exhibit one stable oxidation state, +3 in nearly all cases, thusresembling the lanthanides.

Table 9.2 Oxidation states of the actinides

2+ 3+ 4+ 5+ 6+ 7+

AcTh ?PaUNpPuAm ?Cm ?BkCfEs ?FmMdNoLr

KnownCommon

...



Table 9.3 Electron configurations of the actinides and their ionsa

M M3+ M4+

Ac 6d1 7s2

Th 6d2 7s2 5f1

Pa 5f2 6d1 7s2 5f2 5f1

U 5f3 6d1 7s2 5f3 5f2

Np 5f4 6d1 7s2 5f4 5f3

Pu 5f6 7s2 5f5 5f4

Am 5f7 7s2 5f6 5f5

Cm 5f7 6d1 7s2 5f7 5f6

Bk 5f9 7s2 5f8 5f7

Cf 5f10 7s2 5f9 5f8

Es 5f11 7s2 5f10 5f9

Fm 5f12 7s2 5f11 5f10

Mdb 5f13 7s2 5f12 5f11

Nob 5f14 7s2 5f13 5f12

Lrb 5f14 6d1 7s2 5f14 5f13

aConfigurations are in addition to a radon ‘core’ 1s2 2s2 2p6 3s2 3p6 3d10 4s2 4p6

4d10 4f14 5s2 5p6 5d10 6s2 6p6

bPredicted

Early in the actinide series, electrons in the 6d orbitals are lower in energy than there is5f orbitals, This is clear from the ground-state electronic configurations (Table 9.3) of theatoms, which show that the 6d orbitals are filled before 5f. The 5f orbitals are starting to befilled at protoactinium, and with the exception of curium, the 6d orbitals are not occupiedagain.

The 5f orbitals are not shielded by the filled 6s and 6p subshells as the 4f orbitalsof the lanthanides are (by the corresponding 5s and 5p subshells). Moreover, the energygap between 5f n 7s2 and 5f n−1 6d 7s2 configurations is less than for the correspondinglanthanides. Not only are the 5f orbitals less ‘inner orbitals’ in the sense that the 4f orbitalsare for the lanthanides, and thus more perturbed in bonding, but the near-degeneracy of the5f, 6d, and 7s electrons means that more outer-shell electrons can be involved in compoundformation (and a wider range of oxidation states observed). Thus, the first four ionizationenergies of Th are 587, 1110, 1978, and 2780 kJ mol−1 (Table 9.4) compared with respectivevalues for Zr of 660, 1267, 2218, and 3313 kJ mol−1. So, for the earlier actinides, higheroxidation states are available and, as for the d block, several are often available for eachmetal [relativistic effects may also be important (Section 9.7)]. However, as the 5f electronsdo not shield each other from the nucleus effectively, the energies of the 5f orbitals droprapidly with increasing atomic number, so that the electronic structures of the later actinidesand their ions become more and more like those of the lanthanides, whose chemistry theythus resemble.

9.6 Reduction Potentials of the Actinides

Although a full range of ionization energies is not available throughout the actinide seriesand thus cannot be used predictively, as for the lanthanides (Table 2.2), electrode potentialsare more generally available (Table 9.5) and thus can be so used, as follows:

...


Reduction Potentials of the Actinides 151

Table 9.4 Ionization energies of the actinides (kJ/mol)

I1 I2 I3 I4

Ac 499 1170 1900 4700Th 587 1110 1978 2780Pa 568 1128 2991U 584 1420 1900 3145Np 597 1128 1997 3242Pu 585 1128 2084 3338Am 578 1158 2132 3493Cm 581 1196 2026 3550Bk 601 1186 2152 3434Cf 608 1206 2267 3599Es 630 1216 2334 3734Fm 627 1225 2363 3792Md 635 1235 2470 3840No 642 1254 2643 3956Lr 444 1428 2228 4910

Table 9.5 Reduction potentials of the actinides (V)

M3+ + 3e → M M4+ + 4e → M M3+ + e → M2+ M4+ + e → M3+

Ac −2.13 −4.9Th −1.83 −4.9 −3.7Pa −1.47 −4.7 −2U −1.8 −1.38 −4.7 −0.63Np −1.79 −1.30 −4.7 0.15Pu −2.03 −1.25 −3.5 0.98Am −2.07 −0.90 −2.3 2.3Cm −2.06 −3.7 3.1Bk −1.96 −2.8 1.64Cf −1.91 −1.6 3.2Es −1.98 −1.6 4.5Fm −2.07 −1.1 4.9Md −1.74 −0.15 5.4No −1.26 1.45 6.5Lr −2.1 7.9

1. The negative reduction potentials for the M3+/M potentials indicate that the process

M(s) + 3H+(aq) → M3+(aq) + 3/2 H2(g)

is energetically favourable, suggesting that the metals should react well with dilute acid(and possibly water too), as is in fact observed.

2. The large negative reduction potentials for the M3+/M2+ process early in the seriesindicate that for these metals no aqueous chemistry of M2+ ions is to be expected. Thepotentials become less negative with increasing atomic number, indicating increasingstability of M2+, and the positive value for No is in keeping with No2+ being the moststable ion for this metal in aqueous solution (Table 9.2). A number of compounds havebeen isolated in the solid state for Am2+ (5f 7), Cf2+, and Es2+ (as might be expectedfrom the reduction potentials) and the (+2) ion is known in solution for Fm and Md too(Table 9.2).

...



3. The large negative values of Eo for M4+/M3+ for Th and Pa indicate that reduction toform (+3) species for these metals will be difficult, whilst for U, Np, and Pu the smallerEo values indicate that both the +3 and +4 states will have reasonable stability. However,from Am onwards, Eo > 2V for all these elements (except Bk), suggesting that for allthese metals the (+3) state will be more favoured (as observed). The tendency for the Eo

values for both M4+/M3+and M3+/M2+ to become more positive, and for the reductionto be more favoured, on crossing the series from left to right, shows that overall the loweroxidation states become more stable the higher the atomic number.

9.7 Relativistic Effects

For the lighter chemical elements, the velocity of the electrons is negligible compared withthe velocity of light. However, for the actinides and to a lesser extent the lanthanides this isnot the case; as the velocity of the electrons increases towards c, then their mass increasestoo.

For a 1s electron in a uranium atom, the average radial velocity, Vrad = 92c/137, ∼0.67c.The mass increase is given by m = me/

√(1 − 0.672 = 1.35 me. This produces a contraction

of the 1s orbital and stabilization of 1s electrons. Electrons in other s shells also tend tobe stabilized owing to their orthogonality with 1s. Similar but smaller effects occur for pelectrons. In contrast, d and f electrons tend to be expanded and destabilized (compared withimaginary, ‘non-relativistic’ atoms), due to the increased shielding of the nucleus by the(increasingly stabilized) outer core s and p electrons. Since f electrons are poor at screeningother electrons from the nuclear charge, then as the 4f shell is filled, there is a contractionof the 5p and 6s orbitals.

A important ‘relativistic effect’ is that 5f orbitals of actinides are larger and their electronsmore weakly bound than predicted by non-relativistic calculations, hence the 5f electronsare more chemically ‘available’. This leads to:

(a) a bigger range of oxidation states than with the lanthanides;(b) a greater tendency to covalent bond formation (but maybe involving 6d rather than 5f

orbitals) in ions like MO2+ and MO2

2+(most notably the uranyl ion, UO22+).

Question 9.1 4.5 × 109 tonnes of uranium is present in the Earth’s oceans, at a concentrationof 3.3 mg m−3 (about 1/1000th that in the crust). How might it be extracted?Answer 9.1 Ion-exchange technology is a possibility. The main cost would be that ofpumping all the water needed; wave energy technology (as part of a power station?) couldbe an answer. Recovering other valuable metals (e.g. gold) at the same time might alsoimprove the economics.

Question 9.2 What are the advantages of using UF6for isotope separation?Answer 9.2 It vapourizes at low temperatures, so that little energy is used for that; it hasa low molecular mass, so that, since separation factors are proportional to the differencein mass between the 235U- and 238U-containing molecules, easier separations are achieved;since fluorine is monoisotopic (19F), only molecules of two different masses are involved,minimizing overlap between 235U- and 238U-containing species (this would not be the caseif, say, Cl or Br were involved).

...


Relativistic Effects 153

Question 9.3 What are the patterns in oxidation states seen in (a) the early actinides Ac–Puand (b) the later ones, Am–Lr. With which parts of the Periodic Table are resemblancesmost pronounced? How does this relate to the electronic structure?Answer 9.3 (a) The maximum oxidation state observed for Ac–Np corresponds to thenumber of ‘outer-shell’ electrons. Similar behaviour is seen in transition-metal chemistry,for example for the metals Sc–Mn in the 3d series.

(b) Especially from Bk onwards, the elements exhibit one common oxidation state, +3in almost all cases, resembling the lanthanides in that respect.

...


154

SPH SPH


10 Binary Compoundsof the Actinides


� suggest syntheses for typical compounds;� recognize patterns in formulae and explain them;� recognize patterns in the stability of halides;� recognize and explain trends in the volatility and coordination number in uranium halides;� state why the properties of UF6 make it suitable for isotope separation;� suggest properties for compounds;� appreciate problems in the synthesis of halides of the heavier actinides;� understand the structures of the actinide halides.

10.1 Introduction

This chapter examines some of the most important binary compounds of the actinides,especially the halides. Despite the problems caused by their radioactivity, some binarycompounds of most of these elements have been studied in considerable detail, and form agood vehicle for understanding trends in the actinide series.

The actinides are reactive metals, typified by the reactions of thorium and uranium shownin Figures 10.1 and 10.2. These binary compounds frequently have useful properties. Thus,choosing examples from thorium chemistry, Th2S3 is a high-temperature crucible material,ThN is a superconductor, and Th3P4 and Th3As4 are semiconductors.

10.2 Halides

The compounds known are summarized in Table 10.1. The only compound of an earlyactinide in the +2 state is ThI2, a metallic conductor which is probably Th4+ (e−)2 (I−)2.Certain heavier actinides form MX2 (Am, Cf, Es), which usually have the structure ofthe corresponding EuX2 and are thus genuine M2+ compounds. All four trihalides existfor all the actinides as far as Es, except for thorium and protactinium. Tetrafluorides existfor Th–Cm and the other tetrahalides as far as NpX4 (and in the gas phase in the case ofPuCl4). Pentahalides are only known for Pa, U, and Np; whilst there are a few MF6 (M =U–Pu), uranium is the only actinide to form a hexachloride. The known actinide halides aregenerally stable compounds; most are soluble in (and hydrolysed by) water.


SPH SPH


156 Binary Compounds of the Actinides

300–400°C

400°C

450°C

250°C

600°C

C(excess)

heat

heat heatair,

H2

B

S8

X2

N2

O2

Th

ThO2

ThH2

Th2S5

ThB6,ThB12

ThO2,ThN,Th3N4

ThX4

(X = F, Cl, Br, I)

ThN,Th3N4

ThC2

Figure 10.1Reactions of thorium.

600−1000°C

heat

H2

B

S8

X2

N2

U

UN

600°C 250°C

UH3

UB2, UB4, UB12

800°C

CUC, UC2

P4

500°CU2S3, US2

U3P4

heatUF6, UClx (x = 4–6)

UBr4, UI4, UI3

HXheat

UF4, UX3 (X = Cl, Br, I)

O2

700°C

U3O8

Figure 10.2Reactions of Uranium.

The maximum oxidation state found in compounds of the early actinides correspondsto the total number of electrons available in the 7s, 6d, and 5f orbitals. Beyond uraniumthis is no longer the case, and the stability of the high oxidation states decreases sharply;this probably reflects the decreasing availability of the 5f electrons for bonding. By thesecond half of the series, one oxidation state (+3) dominates, in the way observed forthe lanthanides. Fluorine, the strongest oxidizing agent among the halogens, supports thehighest oxidation states, as is generally observed elsewhere.

10.2.1 Syntheses of the halides

A considerable number of synthetic routes are available, many involving reaction of theoxides with HX or X2(according to the oxidation state desired) though sometimes moreunusual halogenating agents like AlX3 or hexachloropropene have been used. A selectionof methods follow.

SPH SPH


Halides 157

Table 10.1 Oxidation states and halides of the actinides

Oxidation state 2 3 4 5 6

Ac F Cl Br ITh I I F Cl Br IPa I F Cl Br I F Cl Br IU F Cl Br I F Cl Br I F Cl Br F ClNp F Cl Br I F Cl Br I F FPu F Cl Br I F FAm Cl Br I F Cl Br I F F?Cm F Cl Br I F F?Bk F Cl Br I FCf Cl Br I F Cl Br I FEs Cl Br I F Cl Br I F?FmMdNoLr

Fluorides

Ac(OH)3 → AcF3 (HF, 700 ◦C)

AmO2 → AmF3 (HF, heat)

PaO2 → PaF4 (HF/H2, 600 ◦C)

PaF4 → PaF5 (F2, 800 ◦C)

In some cases, where the +3 state is easily oxidized, a reductive route is used:

MO2 → MF3 (M = Pu, Np; HF/H2, 500 ◦C)

UF4 + Al → UF3 + AlF (Unstable) (900 ◦C)

Chlorides

In the case of AnCl3, the route used for lanthanide trihalides involving heating the oxide orhydrated chloride with ammonium chloride works well:

Ac2O3 → AcCl3 (NH4Cl, 250 ◦C)

AmCl3.6H2O → AmCl3 (NH4Cl, 250 ◦C)

NpO2 → NpCl4 (CCl4, 500 ◦C)

Again, reductive methods are sometimes necessary:

UH3 → UCl3 (HCl, 350 ◦C)

NpCl4 → NpCl3 (H2, 600 ◦C)

Bromides

Ac2O3 → AcBr3 (AlBr3, 750 ◦C)

NpO2 → NpBr4 (AlBr3, heat)

M2O3 → MBr3 (M = Cm, Cf; HBr, 600–800 ◦C)

SPH SPH



Table 10.2 Coordination number and structure types of the actinide trihalides

C.N. Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr

Fluorides 11 LaF3 LaF3 LaF3 LaF3 LaF3 LaF3 LaF3 LaF3

9 YF3 YF3

Chlorides 9 UCl3 UCl3 UCl3 UCl3 UCl3 UCl3 UCl3 UCl3

8 PuBr3 PuBr3 PuBr3

6Bromides 9 UCl3 UCl3 UCl3

8 PuBr3 PuBr3 PuBr3 PuBr3 PuBr3

6 AlCl3 AlCl3 AlCl3

6 BiI3 BiI3

Iodides 8 PuBr3 PuBr3 PuBr3 PuBr3

6 BiI3 BiI3 BiI3 BiI3 BiI3

Iodides

Th → ThI4 (I2, 400 ◦C)

CfCl3 → CfI3 (HI, 540 ◦C)

CfI3 → CfI2 (H2, 520–570 ◦C)

10.2.2 Structure Types

Trihalides are known for most of the actinides and form the basis for comparison with thelanthanides. Table 10.2 lists the known structures adopted by the trihalides.

Of these structure types, the LaF3 structure has a fully capped trigonal prismatic struc-ture with 11 coordination (9 + 2); the YF3 structure has rather distorted tricapped trigonalprismatic 9 coordination; the UCl3 structure is a tricapped trigonal prism, from which the8-coordinate PuBr3 structure is derived by removing one of the face-capping halogens; theAlCl3 and BiI3 structures both have octahedral six-coordination.

The actinide trihalides display a similar pattern of structure to those of the lanthanidetrihalides. However, comparing the coordination numbers for Ln3+ and An3+ ions with thesame number of f electrons (‘above one another in the Periodic Table’), it can be seen thatthe coordination number of the lanthanide halides decreases sooner than in the actinideseries, a reflection of the fact that the larger actinide ions allow more halide ions to packaround them. Table 10.3 gives comparative coordination numbers for the trihalides of thelanthanides and actinides.

The tetrahalides have coordination numbers of 8 in the solid state whilst the pentahalideshave 6 and 7 coordination. Thus PaF5 and PaCl5 are both seven coordinate (Figure 10.3),

Table 10.3 Comparison of coordination for trihalide structures


LnF3 11 11 11 11 11 11, 9 11, 9 9 9 9 9 9 9 9 9LnCl3 9 9 9 9 9 9 9 9 8 6 6 6 6 6 6LnBr3 9 9 9 8 8 8 8 6 6 6 6 6 6 6 6LnI3 8 8 8 8 8 6 6 6 6 6 6 6 6 6

Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr

AnF3 11 11 11 11 11 11 11, 9 11, 9AnCl3 9 9 9 9 9 9 9, 8 9, 8 8AnBr3 9 9 9, 8 8 8 8 8, 6 6 6AnI3 8 8 8 8 6 6 6 6 6

SPH SPH


Thorium Halides 159

chlorine=

protactinium=

Figure 10.3The chain structure adopted by PaCl5.

whilst PaBr5 is six coordinate; in contrast UF5 is seven coordinate (as is NpF5), whilst UBr5

and UCl5 are both six coordinate. Hexafluorides are formed by U–Pu (and just possiblyAm), UCl6 is the only hexachloride; all MX6 are octahedral.

The properties and structures of the halides are mainly exemplified in this chapter by aconsideration of the halides of thorium and uranium.

10.3 Thorium Halides

Apart from two rather ill-defined iodides in the +2 and +3 oxidation states, thorium formsexclusively halides in the +4 state. Heating thorium metal with ThI4 affords ThI2 and ThI 3,the latter not well characterized. ThI2 exists in black and gold forms; the high conductivitysuggests that, like some of the apparent lanthanide(II) iodides (Section 3.3.3), it containsfree electrons and is in fact Th4+ (e−)2 (I−)2.

The ThIV halides can be made by various routes including:

ThO2 + 4 HF → ThF4 + 2 H2O

ThH2 + 3 Cl2 → ThCl4 + 2 HCl

Th + 2 Br2 → ThBr4 (700 ◦C)

Th + 2 I2 → ThI4 (400 ◦C)

All are white solids. Table 10.4 summarizes the structures of the thorium(IV) halides (thechloride and the bromide exist in two different forms in the solid state).

Thorium halides form many complexes with neutral donors, discussed in chapter 11.Here it may be noted that there are a number of anionic complexes, including Cs2ThCl6,(pyH)2ThBr6, and (Bu4N)2ThI6, which generally seem to contain isolated [ThX6]2−

Table 10.4 Thorium(IV) halides

Solid state Gas phaseCoord. No. Bond length (A) Closest polyhedron Coord. No. Bond length (A)

ThF4 8 2.30–2.37 Dodecahedron 4 2.14ThCl4 8 2.72–2.90 (β); 2.85–2.89 (α) Sq. antiprism 4 2.58ThBr4 8 2.85–3.12 (β); 2.91–3.02 (α) Sq. antiprism 4 2.72ThI4 8 3.13–3.29 Dodecahedron

SPH SPH



octahedra. However, there is a wide selection of complexes of the smaller fluoride ion,which tend to have higher coordination numbers. The stoichiometry is no guide to thecoordination number – K5ThF9 contains dodecahedral [ThF8]4− ions whilst in Li3ThF7 and(NH4)3ThF7 thorium is nine coordinate

10.4 Uranium Halides

These form a large and interesting series of substances, in oxidation states ranging form+3 to +6, illustrating principles of structure and property. The structures of the uraniumhalides are shown in Figure 10.4.

Oxidationstate

6

5

4

3

Tricappedtrigonalprism

Tricappedtrigonalprism

Bicappedtrigonalprism

Octahedron(chain)

Pentagonalbipyramid

DodecahedronSq. antiprism

Fully cappedtrigonalprism

UF3 UCl3 UBr3 UI3

UI4UBr4UCl4UF4

Oct. dimer Oct. dimer

UBr5

UCl6UF6

ClF Br I

UCl5UF5 β−α−

OctahedronOctahedron

Octahedron Pent-bipy

—

—

—

Figure 10.4Coordination polyhedra in the uranium halides [adapted from J.C. Taylor, Coord. Chem. Rev. 1976, 20, 203; reprintedwith permission of Elsevier Science publishers].

SPH SPH


Uranium Halides 161

10.4.1 Uranium(VI) Compounds

UF6 and UCl6 are two remarkable substances, the former on account of its application inisotopic enrichment of uranium (Sections 9.4 and 10.4.5); the latter as the only actinidehexachloride. Colourless UF6 can be made by a wide variety of routes, including:

UF4 + F2 → UF6 (250–400 ◦C)

3 UF4 + 2ClF3 → 3UF6 + Cl22 UF4 + O2 → UF6 + UO2F2 (600–900 ◦C)

The latter is the Fluorox process, which has the considerable advantage of not requiring theuse of elemental fluorine. Dark green, hygroscopic, UCl6 may be synthesized by halogenexchange or disproportionation:

UF6 + 2 BCl3 → UCl6 + 2 BF3 (–107 ◦C)

2 UCl5 → UCl4 + UCl6 (120 ◦C)

Both have octahedral molecular structures (U–F = 1.994 A in UF6) and are volatile underreduced pressure at temperatures below 100 ◦C.

10.4.2 Uranium(V) Compounds

These are among the best characterized compounds in this somewhat rare oxidation state,although they do tend to be unstable (see the preparation of UCl6) and UI5 does not exist.

2 UF6 + 2 HBr → 2 UF5 + Br2 + 2 HF

2 UO3 + 6 CCl4 → 2 UCl5 + 6 COCl2 + Cl2 (20 atm., 160 ◦C)

2 UBr4 + Br2 → 2 UBr5 (Soxhlet extraction; brown solid)

Grey UF5 has a polymeric structure with 6- and 7-coordinate uranium, whilst red-brownUCl5 and brown UBr5 have a dimeric structure in which two octahedra share an edge.

10.4.3 Uranium (IV) Compounds

In normal laboratory work, these are the most important uranium halides, especially UCl4.

UO2 + 4 HF → UF4 + 2 H2O (550 ◦C)

UO2 + 2 CCl4 → UCl4 + 2 COCl2 (250 ◦C)

U3O8 + C3Cl6 → 3 UCl4 + Cl2C=CClCOCl etc. (reflux)

U + 2 Br2 → UBr4 (He/Br2; 650 ◦C)

U + 2 I2 → UI4 (I2; 500 ◦C, 20 kPa)

HF is used in the synthesis of the tetrafluoride since obviously the use of fluorine in thisreaction would tend to produce UF6. Although Peligot first prepared UCl4 in 1842 by thereaction of uranium oxide with chlorine and charcoal, nowadays it is conveniently made byrefluxing the oxide with organochlorine compounds such as hexachloropropene and CCl4.U4+ does not have reducing tendencies, and UI4 is stable, though not to hydrolysis.

Table 10.5 lists some properties of these compounds. As usual in the lower oxidationstates, the fluorides have significantly lower volatilities. This can alternatively be explained

SPH SPH


Tabl

e10

.5U

rani

um(I

V)

halid

es

UF 4

UC

l 4U

Br 4

UI 4

Des

crip

tion

Air

-sta

ble

gree

nso

lidD

eliq

uesc

entg

reen

solid

Del

ique

scen

tbro

wn

solid

Del

ique

scen

tbla

ckso

lidM

p(◦ C

)10

3659

051

950

6B

p(◦ C

)78

976

1So

lubi

lity

H2O

and

mos

torg

.sol

vent

sH

2O

and

mos

torg

.sol

vent

sH

2O

and

mos

torg

.sol

vent

sC

oord

inat

ion

geom

etry

Squa

rean

tipri

smD

odec

ahed

ron

Pent

agon

albi

pyra

mid

Oct

ahed

ron

C.N

.of

uran

ium

88

76

U–X

dist

ance

(A)

2.25

–2.3

22.

64–2

.87

2.61

(ter

m.)

2.78

–2.9

5(b

ridg

e)2.

92(t

erm

.)3.

08–3

.11

(bri

dge)

162

SPH SPH


Uranium Halides 163

in terms of high lattice energies, on account of the small size of the fluoride ion, or by thegreater ionic character in the bonding. In fact, like the other MF4 and MCl4, UF4 vapourizesas MF4 molecules. As usual, the coordination number C.N. of the metal decreases as thehalogen gets larger whilst the bond lengths increase.

10.4.4 Uranium(III) Compounds

Because of the ease of oxidation of the U3+ ion, these are all made under reducing conditions.

2 UF4 + H2 → 2 UF3 + 2 HF (950 ◦C)

2 UH3 + 6 HCl → 2 UCl3 + 3 H2 (350 ◦C)

2 UH3 + 6 HBr → 2 UBr3 + 3 H2 (300 ◦C)

2 UH3 + 3 HI → 2 UI3 + 3 H2 (300 ◦C)

These compounds have structures typical of the actinide trihalides (Section 10.2.2). GreenUF3 has the 11-coordinate LaF3 structure whilst UCl3 and UBr3, both red, have the tricappedtrigonal prismatic ‘UCl3’ structure. Nine iodide ions cannot pack round uranium, so blackUI3 adopts the 8-coordinate PuBr3 structure (Figure 10.5).

Pu at y = 0

Br at y = 0

Pu at y = ± 12

Br at y = ± 12

Figure 10.5The layer structure of PuBr3 [after A.F. Wells, Structural Inorganic Chemistry, Clarendon Press,Oxford (3rd edn, 1962) and reproduced by permission of the Oxford University Press]. Note that9-coordination is prevented by non-bonding Br . . . Br interaction. The notation y = 0 and y = ± 1

2indicates the relative height of atoms with respect to the plane of the paper. The planes of the layersare normal to the plane of the paper.

10.4.5 Uranium Hexafluoride and Isotope Separation235U is the only naturally occurring fissionable nucleus, but it makes up only 0.72% ofnatural uranium. Scientists working on the Manhattan project to build the first atomicbombs in the early 1940s were faced with the problem of producing uranium enriched in235U. In principle various routes were available, such as deflection in electric and magneticfields, diffusion across thermal or osmotic pressure barriers, and ultracentrifugation. E.O.Lawrence developed a modified cyclotron (‘calutron’) that relied on the fact that 235UCl4+

SPH SPH



and 238UCl4+ ions would follow slightly different paths in a magnetic field and thus beseparated. Although eventually gaseous diffusion became the preferred route, the calutrondid produced significant amounts of enriched uranium by 1945 that were used to makethe first atomic bomb. The preferred technique was eventually gaseous diffusion of UF6

through porous membranes, usually of nickel or aluminium, with a pore size ca. 10–25 nm.Graham’s law of diffusion indicates that the rate of diffusion of a compound ∝ √

1/Mr. Aseparation factor, α*, can be defined as:

α∗ = √Mr(

238UF6)/Mr(235UF6) = √

352/349 = 1.00429.

Using a ‘cascade’ process with up to 3000 stages, an enrichment of up to 90% is obtained.Desirable qualities for the compound used included high volatility, so that it was easy

to convert into a vapour and minimize energy requirements; as low an Mr as possible, inorder to enhance the separation factor; ease of synthesis. UF6 met these criteria quite well;in addition, fluorine has only one isotope, so that the UF6 molecules of any given uraniumisotope all have the same mass and this does not complicate the separation. Although UF6 isactually a solid at room temperature and atmospheric pressure, it is by far the most volatileuranium compound (sublimes at 56.5 ◦C at atmospheric pressure), having a vapour pressureat room temperature of about 120 mm Hg. The principal problem was its high reactivity(many coordination compounds and organometallics were investigated unsuccessfully aspossible alternatives). This meant that fluorinated materials had to be developed for valvesand lubricants (UF6 attacked grease); new pump seals were invented that were gas-tightand greaseless (after the war, the seal material came to be known as Teflon). Fluorine-resistant metals were needed for the porous barriers whilst a high quality of engineeringwas needed to obtain consistent pore sizes. The process was carried out on a huge scale(acres of pores) so a vast plant was needed and the energy requirements for pumping thevapour were immense.

Other methods have been used to carry out uranium enrichment, again using UF6.Gas centrifuges were developed in the 1960s. On centrifugation, the heavier 238UF6

molecules are pushed out more by centrifugal force and tend to accumulate in the pe-riphery, with an excess of 235UF6 in the axial position. Better separations are obtainedat lower temperatures and higher speeds. Commercial plants have used series of vacuumtubes containing 2 metre rotors which are spun at 50 000 to 70 000 rpm. Like gaseousdiffusion, it runs as a cascade process, but requiring many fewer stages than gaseousdiffusion.

Another technique that shows great promise and is still being developed is laser enrich-ment, which can be applied to either atoms or molecules (Figure 10.6). The molecularlaser process exploits the fact that 238UF6 and 235UF6 molecules have slightly differentvibrational frequencies (of the order of 0.5 cm−1). The 235UF6 molecules in UF6 vapour(supercooled in order to produce sharper absorption bands) are selectively excited with atuneable IR laser, then irradiated with a high-intensity UV laser, whereupon the excited235UF6 molecules are photodecomposed into 235UF5 (the 238UF6 molecules are unaffected).Under these conditions, the 235UF5 is a solid, and is separated from the 238UF6 (using a sonicimpactor).

The leading process of this type is the Australian SILEX system (which is also beingused to carry out isotopic separation for other elements such as silicon and zirconium).

SPH SPH


The Actinide Halides (Ac–Am) excluding U and Th 165

Sonicimpactor

235UF5

(solid)

238UF6

235UF6 238UF6

Carrier gas UVlaser

(gas)

IRlaser

Figure 10.6Laser separation method for 235UF6 and 238UF6 (reproduced with permission from S.A. Cotton,Lanthanides and Actinides, Macmillan, 1991, p. 95).

10.5 The Actinide Halides (Ac–Am) excluding U and Th

10.5.1 Actinium

Actinium forms halides only in the (+3) oxidation state, AcX3 and AcOX (X = F, Cl, Br).The trihalides have the LaF3 (X = F) and UCl3 (X = Cl, Br) structures, respectively.

10.5.2 Protactinium

Protactinium has a wealth of halides, with all PaX4 and PaX5 known, also PaI3. There arealso several oxyhalides.

Pa2O5 is the usual starting material for the synthesis of the halides:

Pa2O5 → PaCl5 (C/Cl2/CCl4; 400–500 ◦C, in vacuo)

PaCl5 → PaCl4 (H2; 800 ◦C or Al; 400 ◦C)

Pa2O5 → PaF4 (HF/H2; 500 ◦C)

PaF4 → PaF5 (F2; 700 ◦C)

Pa2O5 → PaBr5 (C/Br2; 600–700 ◦C)

PaBr5 → PaBr4 (Al or H2; 400 ◦C in vacuo)

Pa2O5 → PaI5 (SiI4; 600 ◦C, in vacuo)

PaI5 → PaI4 (Al or H2; 400 ◦C, sealed tube)

PaI5 → PaI3 (360 ◦C, in vacuo)

Seven coordination is found in both colourless PaF5(β-UF5 structure) and yellowPaCl5 (Figure 10.3; chain structure), whilst two forms of dark red PaBr5 adopt the α-and β-UCl5 structures, dimeric with six-coordinate protactinium. PaX4(X = F, Cl, Br)have the eight-coordinate UF4 (X = F; square antiprismatic) and UCl4 (X = Cl, Br;dodecahedral) structures, all eight coordinate; the fluoride is brown, the chloride green-yellow, and the bromide orange-red). Dark brown PaI3 is also eight coordinate (PuBr3

structure). Protactinium shares the same ability to form oxyhalides as several other

SPH SPH



actinides (Ac, Th, U) and transition metals (Nb, Ta, Mo, W).

PaF5 → PaO2F (air; 250 ◦C)

PaCl4 → PaOCl2 (Sb2O3; 150–200 ◦C, in vacuo)

Pa2O5 + PaBr5 → PaOBr3 (sealed tube; 400 ◦C)

PaI5 → PaOI3 and PaO2I (Sb2O3; 150–200 ◦C, in vacuo)

PaOBr3 has a cross-linked chain structure with 7-coordinate protactinium, whilst yellow-green PaOCl2 has a complicated chain structure involving 7-, 8- and 9- coordination, adoptedby several MOX2(M = Th, Pa, U, Np; X = Cl, Br, I).

10.5.3 Neptunium

Neptunium maintains the flavour of a range of halides in different oxidation states, butonly fluorides are known in oxidation states above +4. All NpX3 are known, with assortedpleasant shades (purple fluoride, green chloride and bromide, brown iodide), together withNpX4(X = F, Cl, Br; green, red-orange, and dark red, respectively), NpF5 (blue-white) andNpF6 (orange). Attempts to make the plausible NpF7 using fluorinating agents like NpF5

have not met with success. NpIV has a wide chemistry, but there is no NpI4, though herethere are parallels with the inability of iodide to coexist with a stable ‘high’ oxidation state(cf. Cu2+, Fe3+ given the instability of the known FeI3).

NpF3, NpF4, NpO2 → NpF6 (F2; 500 ◦C)

NpF4 → NpF5 (KrF2/HF)

NpO2 → NpF4 (HF; 500 ◦C)

NpO2 → NpCl4 (CCl4; 500 ◦C)

NpO2 → NpBr4 (AlBr3; 350 ◦C)

NpO2 → NpF3 (H2/HF; 500 ◦C)

NpCl4 → NpCl3 (H2; 450 ◦C)

NpO2 → NpBr3 (HBr; 500 ◦C)

NpO2 → NpI3 (HI; 500 ◦C)

These have the same structures as the uranium analogues, except that a second formof NpBr3 has the PuBr3 structure. Like UF6 and PuF6, NpF6 forms a volatile, toxicvapour (bp 55.2 ◦C). Oxyhalides NpOF3, NpO2F2, NpOF4, NpOCl, and NpOI have beendescribed.

10.5.4 Plutonium

The tendency towards lower stability of high oxidation state noted with Np is continuedhere. Though all PuX3 exist, pale brown PuF4 is the only tetrahalide, and there are nopentahalides, just red-brown PuF6. The trihalides exhibit pastel colours (fluoride violet-blue; chloride blue-green; bromide light green; iodide bright green). The (+2) state is notaccessible; with the next actinide, americium, the reaction with mercury(II) iodide yieldsAmI2.

2 PuO2 + H2 + 6HF → 2PuF3 + 4H2O (600◦C)

Pu2(C2O4)3.10H2O + 6HCl → 2PuCl3 + 3CO + 3CO2 + 13H2O

SPH SPH


The Actinide Halides (Ac–Am) excluding U and Th 167

2PuH3 + 6HBr → 2PuBr3 + 3H2 (600◦C)

2Pu + 3HgI2 → 2PuI3 + 3Hg (500◦C)

Pu4+(aq) + 4HF(aq) → PuF4.21/2H2O + HF(g) → PuF4

PuF4 + F2 → PuF6 (400◦C)

Two further points merit attention. PuF6 is a very volatile substance, like the other actinidehexafluorides (mp 51.5 ◦C; bp 62.15 ◦C); very reactive (it undergoes controlled hydrolysisto PuO2F2 and PuOF4), it tends to be decomposed by its own α-radiation, and is best keptin the gaseous state (resembling NO2 in appearance).

As already noted, PuCl4 does not exist in the solid state; however, it has been detectedin the gas phase above 900 ◦C by its UV–visible absorption spectrum, so the equilibriumbelow moves to the right on heating:

PuCl3 + 1/2 Cl2 � PuCl4

Halide complexes in the (+4) state are stable for all but iodides. Apart from the twomentioned above, other oxyhalides include PuO2Cl2.6H2O and PuOX (X = F, Cl, Br, I),the absence of any in the (+4) and (+5) states is again notable.

10.5.5 Americium

At americium, dihalides become isolable for the first time.

Am + HgX2 → AmX2 + Hg (X = Cl, Br, I; heat)

They are black solids with the PbCl2 (X = Cl); EuBr2 (X = Br); and EuI2 (X = I) structures.As with plutonium, only fluorides occur in the higher (>3) oxidation states. The trihalidesare important; they have the usual pastel colours, pink in the case of the fluoride and chloride,white–pale yellow (bromide), and pale yellow (iodide).

AmO2 + 4HF (aq) (with HNO3) → AmF4.xH2O

AmF4.xH2O → AmF3 (NH2HF2; 125◦C)

AmO2 → AmCl3 (CCl4; 800◦C)

AmO2 → AmBr3 (AlBr3; 500◦C)

AmO2 → AmI3 (AlI3; 500◦C)

AmF3 + 1/2F2 → AmF4

There is an unsubstantiated report of the isolation of AmF6 (Iu. V. Drobyshevshii et al.,Radiokhimiya, 1980, 22, 591):

AmF4 → AmF6 (KrF2/HF; 40–60 ◦C)

The product was a volatile dark brown solid, with an IR absorption at 604 cm−1 (compareν U–F at 624 cm−1 in UF6).

SPH SPH



10.6 Halides of the Heavier Transactinides

Earlier mention (Section 10.2.2 and Tables 10.1–10.3) of the trihalides of the transplutoniumelements does not do justice to the synthetic expertise of experimentalists who have studiedthese compounds. Some examples of the techniques involved follow.

10.6.1 Curium(III) Chloride

The first synthesis, reported in the mid 1960s, was made difficult by the use of the short-lived 244Cm (t1/2 = 18.1 y) isotope, which caused significant damage to the sample through

irradiation. However, by 1970 microgram quantities of 248Cm (t1/2 = 3.4 × 105 y) wereavailable, albeit on a microgram scale. Ion-exchange beads were loaded with 1 µg of248Cm, then heated to 1200 ◦C to convert the curium into the oxide; the Cm2O3 was placedin capillaries and chlorinated using HCl at 500 ◦C, forming pale yellow CmCl3. Slow coolingof molten CmCl3 from over 600 ◦C afforded a single crystal of the curium chloride, whichwhen studied by X-ray diffraction revealed its adoption of the UCl3 structure and affordedthe Cm–Cl bond lengths (J. R. Peterson and J. H. Burns, J. Inorg. Nucl. Chem., 1973, 35,1525).

10.6.2 Californium(III) Chloride, Californium(III) Iodide, and Californium(II) Iodide

CfCl3 was first reported by Burns et al., in 1973, synthesized by the ion-exchange bead routedescribed above. As more 249Cf (t1/2 = 351 y) became available on the µg scale, anotherroute was employed. A precipitate of californium oxalate was filtered in a quartz capillaryusing a piece of ashless filter paper. Heating at 800–900 ◦C destroyed the filter paper andturned the oxalate into Cf2O3, which reacted with CCl4 at 520–550 ◦C to form lime-greenCfCl3. X-ray diffraction showed that this had the UCl3 structure. The CfCl3 was purifiedby sublimation at 800 ◦C, then employed in the synthesis of two iodides:

CfCl3 → CfI3 (HI; 540 ◦C)

The yellow-orange triiodide forms the lavender-violet diiodide on reduction:

CfI3 → CfI2 (H2; 520–570 ◦C)

These compounds were identified using X-ray diffraction and UV-visible spectroscopy.

10.6.3 Einsteinium(III) Chloride

The first synthesis of this compound was carried out on the nanogram scale in 1968 byPeterson’s group in the face of several difficulties. The 253Es isotope is so short lived(t1/2 = 20 d) that it releases some 1400 kJ mol−1 min−1, sufficient to dislocate the crystallattice within 3 minutes of formation and it is also capable of destroying ion-exchangeresins. Around 200 nanograms of einsteinium were absorbed by a charcoal chip, which washeated to turn the Es into Es2O3, then chlorinated using CCl4 to form EsCl3. A crystal ofEsCl3 was just heated to below its melting point, so that defects in the lattice produced bythe radiation were continuously annealed out. One decade later, in 1978, a different routewas employed by Peterson to make Es2O3, first precipitating the oxalate, then converting itinto the oxide by calcination. Treatment of the oxide first with HF, then fluorine at 300 ◦C,

SPH SPH


Oxides 169

yielded the fluoride EsF3. The absorption spectrum of EsF3 was examined over a period of2 years, so that additionally the spectra of the 249BkF3 daughter and 249CfF3 granddaughterwere also obtained, enabling their assignment of the LaF3 structure.

10.7 Oxides

Table 10.6 summarizes the formulae of the oxides. It should be emphasized that this repre-sents a simplification and, especially for U and Pu, a number of phases exist between thecompositions shown.

The pattern of oxidation state here is very different to that of the lanthanides, where thecommon oxides are Ln2O3. The early actinides in particular show much more resemblanceto the transition metals, where the maximum oxidation state corresponds to the number of‘outer shell’ electrons; this reflects the greater availability of d and f electrons in the actinideelements. The +4 state in particular is more stable than in the lanthanide series, where it isonly encountered in a few ions.

10.7.1 Thorium Oxide

ThO2, is a white solid that adopts the fluorite structure (as do the MO2 phases of the otheractinides). When heated it gives off a rather bluish light; if about 1% cerium is added,the light is both whiter and more intense so that the mixture came to be used in makingincandescent gas mantles, widely used for lighting until comparatively recently.

10.7.2 Uranium Oxides

The uranium–oxygen phase diagram is very complicated, with some of the 14 reportedphases not being genuine and several phases showing variable composition. The importantphases are UO2, U4O9, U3O8, and UO3. Brown-black UO2 has the fluorite structure. It isbest made by reduction of higher uranium oxides (e.g. UO3 with H2 or CO at 300–600 ◦C).

Additional oxygen can be incorporated into interstitial sites in the basic fluorite structureuntil the composition reaches U4O9(black UO2.25). Green-black U3O8 is the result of heatinguranyl salts at around 650–800 ◦C; above 800 ◦C, it tends to lose oxygen.

3 UO2(NO3)2 → U3O8 + 6 NO2 + 2 O2

This is a mixed-valence compound with pentagonal bipyramidal coordination of uranium.Addition of more oxygen eventually results in orange-yellow UO3. This has several crys-talline forms, most of which contain uranyl groups in 2 + 4 coordination, and can be madeby heating, e.g., (NH4)2U2O7 or UO2(NO3)2 at 400–600 ◦C (above which temperature, oxy-gen loss to U3O8 tends to occur). When uranium oxides are heated with M2CO3 or M’CO3

Table 10.6 Oxides of the actinide elements

Ac Th Pa U Np Pu Am Cm Bk Cf Es

Ac2O3 Pu2O3 Am2O3 Cm2O3 Bk2O3 Cf2O3 Es2O3

ThO2 UO2 NpO2 PuO2 AmO2 CmO2 BkO2 CfO2

Pa2O5 Np2O5

UO3 NpO3 (?)

SPH SPH



(M, M’ are group I or II metals), uranates with formulae such as Na2UO4 or K2U2O7 result;unlike d-block metal compounds with similar formulae, these contain UO6 polyhedra.

10.8 Uranium Hydride UH3

The +3 oxidation state of uranium is strongly reducing, so it is not altogether surprisingto find it forms a stable hydride. Massive uranium reacts with hydrogen gas on heating toaround 250 ◦C (at lower temperatures if powdered metal is used), swelling up to a fine blackpowder, which is pyrophoric in air. It decomposes on further heating (350–400 ◦C under ahydrogen atmosphere) to hydrogen and uranium powder, so that making then decomposingthe hydride can be a useful way of preparing a reactive form of the metal (a strategy thatcan be followed with other actinides, e.g. Th). Some reactions of UH3 are summarized inFigure 10.7. The crystal structure shows that each uranium has 12 hydrogen neighbours.

UH3

HFUF4

HCl UCl3

UBr3

UI3

HBr

HI

Cl2Br2

H2O

O2

H2SNH3

UNUS2

U3O8

UO2

UBr4

UCl4

25°C

20°C

350°C

250°C

250°C

400°C

300°C

300°C

300°C

300°C

Figure 10.7Reactions of UH3.

10.9 Oxyhalides

Although not binary compounds, a discussion of actinide oxyhalides is not out of placehere. A number of these are formed by the earlier actinides, another difference between the4f and 5f metals.

Thorium forms ThOX2 (X = Cl, Br, I) but there is a greater variety with protactinium:

PaF5.2H2O → PaO2F (air; 250 ◦C)

PaCl4 → PaOCl2 (Sb2O3; 150–200 ◦C)

Pa2O5 + PaBr5 → PaOBr3 (sealed tube; 400 ◦C)

PaI5 → PaOI3 and PaO2 I (Sb2O3; 150–200 ◦C)

Of these, yellow PaOBr3 has a cross-linked chain structure with seven-coordinate Pa (boundto 3 O and 4 Br). Yellow-green PaOCl2 has an important structure adopted by many

SPH SPH


Oxyhalides 171

oxyhalides MOX2 (X = Cl, Br, I; M = Th, Pa, U, Np), which contains seven-, eight-,and nine-coordinate Pa.

Neptunium forms NpOF3, NpO2F2, NpOF4, and NpOX (X = Cl, I), whilst the plutoniumcompounds PuOX (X = F, Cl, Br, I), PuO2Cl2.6H2O, PuOF4, and PuO2F2 are known, aswell as AmO2F2 (and certain americium oxychloride complexes).

The most important oxyhalides, though, are the uranium compounds. Yellow UO2F2 andUO2Cl2, and red UO2Br2, have been known for a long while, but the existence of beigeUO2I2 was doubted for many years, only being confirmed very recently (2004) by Berthetet al.

UO3 → UO2F2 (HF; 400 ◦C)

U3O8 → UO2Cl2.2 H2O (HCl/H2O2)

UO2Cl2.2 H2O → UO2Cl2 (Cl2/HCl; 450 ◦C)

UCl4 → UO2Cl2 (O2; 350 ◦C)

UBr4 → UO2Br2 (O2; 170 ◦C)

UO2(OTf)2 → UO2I2 (Me3SiI; OTf = CF3SO3)

In the solid state, UO2F2 has a structure in which a uranyl unit is bound to six fluorides (allbridging); whilst in UO2Cl2 uranium is bound to the two ‘oxo’ oxygens, four chlorines,and another oxygen, from another uranyl unit. UO2I2 has not been obtained in crystallineform, but it is soluble in pyridine and THF (L), forming UO2I2L3, which have pentagonalbipyramidal coordination of uranium, with the usual trans- O=U=O linkage.

A number of adducts of the oxyhalides are well characterized, such as all-trans-[UO2X2L2] [X = Cl, L = Ph3PO; X = Br, L = (Me2N)3PO; X = I, L = (Me2N)3PO,Ph3PO, Ph3AsO]. Others are anionic, containing ions such as [UO2F5]3− and [UO2X4]2−

(X = Cl, Br). Hydrates have recently been reported of the bromide and iodide, all-trans-UO2I2(OH2)2·4Et2O, [UO2Br2(OH2)3] (cis-equatorial bromides), and the dimer[UO2Br2(OH2)2]2, which is [Br(H2O)2O2U(µ-Br)2UO2Br(OH2)2] .

Question 10.1 Discuss the halides formed by the actinides, commenting on the trends inoxidation state as the series is crossed.Answer 10.1 Early on in the series, as far as uranium, the maximum oxidation statecorresponds to the total number of ‘outer shell’ electrons. Uranium forms a hexachloride,in addition to the MF6 also formed by Np and Pu. After uranium, neptunium forms the fullrange of tetrahalides, but, from plutonium onwards, the (+3) state dominates the chemistryof the binary halides, which strongly resemble those of the lanthanides. This may reflectdecreased availability of 5f (and 6d?) electrons for bonding. As usual, F supports the highestoxidation states.

Question 10.2 Study the information about the thorium tetrahalides in Table 10.4 andcomment on the trends and patterns in it.Answer 10.2 The bond lengths increase as the atomic number of the halogen increases,since the atomic (ionic) radius of the halogen increases. The bond length is less for the ThX4

molecules in the gas phase than for the corresponding 8-coordinate species in the solid state;again this is expected since there will be more interatomic repulsion in the eight-coordinatespecies. Since there is no obvious pattern in which compounds adopt particular polyhedra in

SPH SPH



the solid state, this suggests that the dodocahedron and square antiprism have very similarenergies. [It may be parenthetically noted that there is no decrease in coordination numberas the size of the halogen increases, as seen for uranium (Figure 10.4).]

Table 10.7 Melting points and boiling points of uranium chlorides

UCl3 UCl4 UCl5 UCl6

Mp (◦C) 835 590 287 177.5 (dec.)Bp (◦C) dec. 790

Question 10.3 Table 10.7 shows the melting and boiling points of the chlorides of uranium.Comment on how these relates to their structures.Answer 10.3 There is a clear relationship with the degree of molecularity. UCl6 has amonomeric molecular structure with relatively weak intermolecular forces and is thereforethe most volatile (and on account of its high oxidation state, the least stable); UCl5 isdimeric U2Cl10, therefore as a larger molecule will have stronger intermolecular forces andbe less volatile than UCl6. Continuing to UCl4 and UCl3, the more halogens surroundingthe uranium, the more energy will be needed to break the lattice into small mobile units,and the higher will be the melting and boiling points.

Question 10.4 Thorium(IV) oxide has a melting point of 3390 ◦C. Explain why.Answer 10.4 It is composed of Th4+ and O2− ions. The strong electrostatic attractionbetween these highly charged ions (a very high lattice energy) means that a great deal ofenergy has to be supplied to overcome the attraction in order to free the ions and melt it.

Question 10.5 Write equations for the reduction of UO3 by both hydrogen and carbonmonoxide.Answer 10.5 UO3 + H2 → UO2 + H2O; UO3 + CO → UO2 + CO2

Question 10.6 UH3 is high melting and has a high conductivity, similar to uranium metal.Comment on its bonding and structure.Answer 10.6 The high conductivity suggests the presence of delocalized electrons. Onepossible structure is U4+(H−)3(e−); the ionic nature would explain the high melting point.

SPH SPH


11 Coordination Chemistryof the Actinides


� recognize +4 and +6 as the main oxidation states for uranium and +4 as the onlyimportant oxidation state for thorium;

� appreciate that coordination numbers of >6 are the norm;� be familiar with the main features of the chemistry of the other actinides, and appreciate

the transition in chemical behaviour across the series;� appreciate the importance of nitrate complexes;� recognize the characteristic structures of uranyl complexes;� suggest structures for compounds;� explain the separation of nuclear material and the recovery of U and Pu in terms of the

chemistry involved;� appreciate the problems presented in nuclear waste disposal and understand the solutions;� suggest safe ways to handle actinide compounds.

11.1 Introduction

Studies of the coordination chemistry of the actinides have been limited by a number offactors – the care needed in handling radioactive materials and the possibility of damage tohuman tissue from the radiation; toxicity (especially Pu); the very small quantities availableand very short half-lives of the later actinides; radiation and heating damage to solutions;and radiation damage (defects and dislocations) to crystals.

The vast majority of the studies reported have concerned the metals thorium and uranium,particularly the latter, due to accessibility of raw materials, ease of handling, and the longlifetimes of the relatively weakly α-emitting elements Th and U. In many cases, compoundsof neptunium and plutonium with similar formulae to U and Th analogues have been madeand found to be isomorphous and thus presumably isostructural. This chapter will thereforecommence with, and concentrate largely on, the chemistry of complexes of these elements,followed by sections on the other actinides.

11.2 General Patterns in the Coordination Chemistry of the Actinides

As noted earlier (Table 9.2), the patterns of oxidation states of the early actinides resemblethose of d-block metals, with the maximum oxidation state corresponding to the numberof ‘outer shell’ electrons. Thus the chemistry of thorium is essentially confined to the +4


SPH SPH


174 Coordination Chemistry of the Actinides

state (to an even greater degree than Zr or Hf) but uranium exhibits oxidation states of +3,+4, +5, and +6, with most compounds being in either the +4 or +6 state. The reason forthis is that uranium(III) compounds tend to be easy to oxidize:

U4+(aq) + e− → U3+(aq) E = −0.63 V

whilst under aqueous conditions the UO2+ ion readily disproportionates to a mixture of

U4+ and UO22+ [though there is a chemistry of uranium(V) in non-aqueous solvents].

As the actinide series is crossed, it becomes harder to involve the f electrons in com-pound formation, so that the later actinides increasingly show a chemistry in the +3 state,resembling the lanthanides, but with a more prominent +2 state.

11.3 Coordination Numbers in Actinide Complexes

As for the lanthanides, actinide complexes display high coordination numbers. A study ofthe aqua ions of early actinides makes an interesting comparison (Table 11.1 lists numbersof water molecules and bond lengths).

Table 11.1 Actinide aqua ions – numbers of bound water molecules and metal–water distancesa

Ox state Ac Th Pa U Np Pu Am Cm Bk Cf

6 MO 2+2 5.0 5.0 6.0

M–OH2 (A) 2.40 2.42 2.40–2.455 MO +

2 5.0 4.0M–OH2 (A) 2.50 2.47

4 M4+ 10.0 9.0; 10.0 11.2 8 or 9M–OH2 (A) 2.45 2.51; 2.42 2.40 2.39

3 M3+ 9 or 10 9 or 10 10.2 10.3 10.2 8.5 ± 1.5M–OH2 (A) 2.61 2.52 2.51 2.48 2.45 2.4

a The uncertainty in the hydration number is generally in the region of 1.

The best characterized aqua ion is the hydrated uranyl ion [UO2(OH2)5]2+, which hasbeen isolated in several salts studied by diffraction methods as well as X-ray absorptionmethods (EXAFS) in solution. However, similar ions are firmly believed to exist in the casesof [MO2(aq)]2+ (M = Np, Pu) and [MO2(aq)]+ (M = U, Np, Pu). In studying Table 11.1,two features stand out. Ions like UO2

2+ are unprecedented in lanthanide chemistry, both inrespect of the +6 oxidation state and also in the presence of strong and non-labile U=Obonds. Secondly, the coordination numbers of the hydrated +3 actinide ions appear to behigher than those of the Ln3+ ions (9 for early lanthanides, 8 for later ones), explicableon account of the slightly higher ionic radii of the actinide ions. One of the few othercharacterized aqua ions is the trigonal prismatic [Pu(H2O)9]3+, strongly resembling thecorresponding lanthanide series.

11.4 Types of Complex Formed

Table 11.2 gives some stability constants for complexes of Th4+ and UO22+. Thorium

forms stronger complexes with fluoride, the ‘hardest’ halide ion, than with chloride andbromide; this is the behaviour expected of a ‘hard’ Lewis acid. Thorium also forms quite

SPH SPH


Uranium and Thorium Chemistry 175

Table 11.2 Stability constants of Th4+ and UO 2+2 complexes

Stability constants for complexes of Th4+ Stability constants for complexes of UO 2++2

Ligand I (mol/dm3) Log K1 Log K2 Log K3 Log K4 Ligand I (mol/dm3) Log K1

F− 0.5 7.56 5.72 4.42 F− 1 4.541 7.46 Cl− 1 −0.10

Cl− 1 0.18 Br− 1 −0.3Br− 1 −0.13 NO −

3 1 −0.3NO−

3 1 0.67 EDTA4− 0.1 7.4SO 2−

4 2 3.3 2.42NCS− 1 1.08EDTA4− 0.1 25.3acac− 0.1 8.0 7.5 6.0 5.3

strong complexes with oxygen-donor ligands like nitrate. The hexadentate ligand EDTA4−

forms very stable complexes, due largely to the favourable entropy change when six watermolecules are replaced.

11.5 Uranium and Thorium Chemistry

11.5.1 Uranyl Complexes

The great majority of uranium(VI) compounds contain the UO2 group and are known asuranyl compounds; exceptions are a few molecular compounds, such as the halides UOF4,UF6, and UCl6, and some alkoxides such as U(OMe)6. Uranyl compounds result eventuallyfrom exposure of compounds of uranium in other oxidation states to air. They character-istically have a yellow fluorescence under UV light; from the early 19th century, glassmanufacturers added uranium oxide when making yellow and green glass (it is sometimesknown as Vaseline glass).

Uranyl complexes can be thought of as derivatives of the UO22+ ion. There is a very

wide range of them; they may be cationic, such as [UO2(OH2)5]2+ ions; neutral, e.g.[UO2(OPPh3)2Cl2]; or anionic, such as [UO2Cl4]2−, yet all feature a trans-UO2 group-ing with the characteristic short U–O bonds (1.7–1.9 A), quite comparable with those foundin ‘osmyl’ compounds, which have the OsO2 grouping. The presence of the uranyl groupcan readily be detected in the IR spectrum of a uranium compound through the presenceof a strong band in the region 920–980 cm−1 caused by the asymmetric O–U–O stretch-ing vibration; a corresponding band around 860 cm−1 caused by the symmetric O–U–Ostretching vibration is seen in the Raman spectrum. Fine structure due to symmetric uranylstretching vibrations can be seen on an absorption peak in the spectrum of uranyl complexesaround 450 nm (Figure 12.1). There has been considerable speculation on the bonding inthe uranyl ion. The essentially linear geometry of the UO2 unit is an invariable feature ofuranyl complexes; no other atoms can approach the uranium nearer than ∼2.2 A. Uraniumd–p and f–p π bonding have both been invoked to explain the bonding (Figure 11.1).

An energy level diagram for the uranyl ion is shown in Figure 11.2. An electron counttakes 6 electrons from uranium, four from each oxygen, deducting two for the positivecharges; alternatively, if the uranyl ion is thought of as a combination of U6+ and two O2−,taking six electrons from each oxide and none from U6+, again giving 12. These completelyoccupy the six σu, σg, πu, and πg molecular orbitals (σu

2 σg2 πu

4 πg4). It has been suggested

SPH SPH


Figure 11.1π -bonding in the uranyl, [UO2]2+ ion: (a) dxz–px overlap; (b) fxz2 –px overlap; (c) σ -bonding in theuranyl ion (reproduced with permission from Figure 3.24 of S.A. Cotton, Lanthanides and Actinides,Macmillan, 1991).

σg

σu

σu

σg

πu

πg

σg

πgπu

2p

The relative ordering of the bonding MOs is uncertain

Occupiedin UO

22+

σu

σg

φu

πu

πg

δu

δg

5f

6d δgδg

πu

δuφu

πg

Figure 11.2M.O. scheme for the uranyl ion, UO2

2+ (reproduced with permission from Figure 3.25 of S.A. Cotton,Lanthanides and Actinides, Macmillan, 1991).

176

SPH SPH



that repulsion (antibonding overlap) between oxygen p orbitals and occupied uranium 6dand 5f orbitals may destabilize the bonding d molecular orbitals, causing them to be higherin energy than the π bonding orbitals, as shown. Addition of further electrons puts themin the essentially non-bonding δu and φu orbitals, accounting for the existence of the ratherless stable MO2

2+ (M = Np, Pu, Am) ions.The uranium(V) species UO2

+ exists, but is less stable than UO22+, possibly owing to

weaker overlap; it readily decomposes by disproportionation:

2 UO2+(aq) + 4 H+(aq) → UO2

2+(aq) + U4+(aq) + 2 H2O(l)

In contrast to the trans geometry of UO22+, isoelectronic (6d0 5f 0) ThO2 molecules in the

gas phase or in low temperature matrices are bent (∠O–Th–O ∼ 122◦); this is thought to bedue to the thorium 5f orbitals being much higher in energy (in uranium, the 5f orbitals arewell below 6d), reducing possibilities of 5f–p overlap, so that thorium resorts to using 6dorbitals in the p bonding, like transition metals, as in the case of MoO2

2+(5d0) and WO22+

(6d0) ions which also adopt bent, cis geometries (∠O–M–O ∼ 110 ◦).

11.5.2 Coordination Numbers and Geometries in Uranyl Complexes

An extensive range of uranyl complexes has been prepared and had their structures deter-mined. Their structure can be summarized as a uranyl ion surrounded by a ‘girdle’ of 4,5, or 6 donor atoms round its waist (a rare example of 2 + 3 coordination is known for acomplex of the uranyl ion with a calixarene ligand; another is the amide complex [K(thf)2][UO2{N(SiMe3)2}3]). If the ligands are monodentate donors, there are usually 4 of them,unless they are small, like F or NCS, when five can be accommodated. When bidentateligands with small steric demands like NO3, CH3COO, and CO3 can be accommodated,six donor atoms can surround the uranyl group. In general if there are 4 or 5 donor atomsround the waist, they are reasonably coplanar, but puckering sometimes occurs when thereare six. Table 11.3 shows examples of uranium complexes for 5-, 6-, 7-, and 8-coordination.

Table 11.3 Uranyl complexes

5-coordinate (2 + 3) 6-coordinate (2 + 4) 7-coordinate (2 + 5) 8-coordinate (2 + 6)

[UO2{N(SiMe3)2}3]− Cs2[UO2Cl4] UO2Cl2 UO2F2

(Me4N)2[UO2Br4] UO2(superphthalocyanine) UO2CO3

MgUO4 [UO2(NO3)2(Ph3PO)] [UO2(NO3)2(H2O)2]BaUO4 [UO2(L)5]2+ CaUO4

(L, e.g., H2O, DMSO, urea) SrUO4

Rb[UO2(NO3)3]

The uranium(VI) aqua ion is now firmly established as the pentagonal bipyramidal[UO2(OH2)5]2+; it has been found in crystals of the salt [UO2(OH2)5] (ClO4)2.2H2O (U=O1.702 A, U–OH2 2.421 A) and also found in solutions by X-ray diffraction studies. Sim-ilar [UO2(L)5]2+ ions (L = urea, Me2SO, HCONMe2) also exist. Uranyl nitrate formscomplexes with phosphine oxides of the type [UO2(NO3)2(R3PO)2]; similar phosphatecomplexes [UO2(NO3)2{(RO)3PO}2] are important in the extraction of uranium in nuclearwaste processing (Figure 11.3 and Section 11.5.5). A number of structures of these com-plexes are known; for example, when R=isobutyl, U=O 1.757 A; U–O(P) 2.372 A; U–O(N)2.509 A.

SPH SPH



U

O

O

O

(RO)3P

O

OO

O

NO

N O

P(OR)3

O

Figure 11.3Structure of UO2(NO3)2[(RO)3PO]2.

11.5.3 Some Other Complexes

Uranyl carbonate complexes have attracted considerable interest in recent years as they areintermediates in the processing of mixed oxide reactor fuels and in extraction of uraniumfrom certain ores using carbonate leaching; more topically they can be formed when uranylores react with carbonate or bicarbonate ions underground, and can be present in relativelyhigh amounts in groundwaters. The main complex formed in carbonate leaching of uranylores is 8 coordinate [UO2(CO3)3]3−, but around pH 6 a cyclic trimer [(UO2)3(CO3)6]6− hasbeen identified.

UO3 dissolves in acetic acid to form yellow uranyl acetate, UO2(CH3COO)2.2H2O. Itformerly found use in analysis since, in the presence of M2+ (M = Mg or Zn), it precipitatessodium ions as NaM [UO2(CH3COO)3]3.6H2O.

N

N

N

N

N NN

N

N

N

U

O

O

NN N

N NN

NN M

heat

(a)

MX2

(M = Cu, Zn, Co, for example; X = halide)

CN

CN

UO2Cl2 +

Figure 11.4Structure and reaction of “uranyl superphthalocyanine”.

SPH SPH



U

O

O

H2O O

OO

O

NO

N O

OH2

Figure 11.5Structure of UO2(NO3)2(H2O)2.

Uranyl chloride, UO2Cl2, reacts on heating with o-phthalodinitrile to form a so-called ‘superphthalocyanine’ complex with 2 + 5 coordination (Figure 11.4); other metals(lanthanides, Co, Ni, Cu) react with this, forming a conventional phthalocyanine, so thatthe uranyl ion has an important role in sustaining this unusual structure.

11.5.4 Uranyl Nitrate and its Complexes; their Role in Processing Nuclear Waste

Reaction of uranium oxide with nitric acid results in the formation of nitrates UO2(NO3)2.xH2O (x = 2, 3, 6); the value of x depends upon the acid concentration. All con-tain [UO2(NO3)2(H2O)2] molecules; the nitrate groups are bidentate, so that uranium is8 coordinate (Figure 11.5). Its most important property lies in its high solubility in a rangeof organic solvents in addition to water (Table 11.4), which is an important factor in theprocessing of nuclear waste.

Table 11.4 Solubility of uranyl nitrate in various solvents

Solvent Solubility (g per g solvent)

Water 0.540Diethyl ether 0.491Acetone 0.617Ethanol 0.675

By adding metal nitrates as ‘salting out’ agents, the solubility of uranyl nitrate in water canbe decreased to favour its extraction from aqueous solution into the organic layer. Tributylphosphate [TBP; (C4H9O)3P=O] acts as a complexing agent (in the manner discussed inthis chapter for a number of phosphine oxide ligands) and also as solvent, with no salting-out agent being needed. In practice, a solution of TBP in kerosene is used to give betterseparation, as pure TBP is too viscous and also has a rather similar density to that of water.

11.5.5 Nuclear Waste Processing

Nuclear fuel rods consist of uranium oxide pellets contained in zirconium alloy or steeltubes. As the fission process proceeds, uranium is used up and fission products accumulate.A lot of these fission products are good neutron absorbers and reduce the efficiency of thefission process (by absorbing neutrons before they reach uranium atoms) so that the rodsare removed for reprocessing before all the 235U content has undergone fission. Fissionof a 235U atom produces two lighter atoms of approximate relative atomic masses around90–100 and 130–140, with the main fission products being the intensely radioactive andshort lived 131I (t1/2 ∼ 8 d), 140La, 141Ce, 144Pr, 95Zr, 103Ru, and 95Nb, and longer-lived137Cs, 90Sr, and 91Y. These essentially useless and toxic products have to be separated fromunchanged uranium and also from plutonium (the product of neutron absorption by 238U),both of which can be used again as fuels.

SPH SPH



1. The first stage of the process involves immersing the fuel rods in ponds of water forup to 3 months. This allows the majority of the short-lived and intensely radioactive fissionproducts such as 131I to decay.

2. The rods are then dissolved in rather concentrated (7M) nitric acid, producing a mixtureof UO2(NO3)2, Pu(NO3)4, and other metal nitrates.

3. The mixture is extracted with a counter-current of a solution of TBP in kerosene.Uranium and plutonium are extracted into kerosene as the complexes [UO2(NO3)2(tbp)2]and [Pu(NO3)4(tbp)2], but the other nitrates, of metals such as the lanthanides and actinidesbeyond Pu, as well as fission products, do not form strong complexes with TBP and stay inthe aqueous layer.

4. The mixture of uranium and plutonium is treated with a suitable reducing agent [iron(II)sulfamate, hydrazine, or hydroxylamine nitrate]; under these conditions, UVI is not reducedand stays in the kerosene layer, but PuIV is reduced to Pu3+, which is only weakly complexedby TBP and so migrates into the aqueous phase.

UO22+(aq) + 4 H+(aq) + 2 e− → U4+(aq) + 2 H2O (l) E = +0.27 V

Pu4+(aq) + e− → Pu3+(aq) E = +1.00 VFe3+(aq) + e− → Fe2+(aq) E = +0.77 V

The uranyl nitrate is extracted back into the aqueous phase and crystallized as thehydrate UO2(NO3)2.xH2O; thermal decomposition to UO3 is followed by hydrogen re-duction to re-form UO2. The plutonium is reoxidized to Pu4+, precipitated as the oxalatePu(C2O4)2.6H2O, which undergoes thermal decomposition to PuO2.

The fission products remain to be dealt with. The preferred solution at the moment is toevaporate their solution and pyrolyse the product to convert it into a mixture of oxides; onfusion with silica and borax an inert borosilicate glass is formed, which encapsulates theradioactive materials. At present the main problem seems to be choosing suitably stablegeological areas where these materials will not be disturbed by earthquakes nor dissolvedby underground water.

11.6 Complexes of the Actinide(IV) Nitrates and Halides

11.6.1 Thorium Nitrate Complexes

A number of thorium nitrate complexes have been synthesized and studied. Hydrated tho-rium nitrate, Th(NO3)4.5H2O, contains [Th(NO3)4.(H2O)3] molecules and was one of thefirst 11-coordinate compounds to be recognized (Figure 11.6).

Reaction of thorium nitrate with tertiary phosphine oxides in solvents like ace-tone or ethanol has afforded a number of complexes. In particular, when thoriumnitrate reacts with Me3PO, a wide range of complexes is obtained depending uponthe stoichiometry and solvent used, compounds with formulae Th(NO3)4.5Me3PO,Th(NO3)4.4Me3PO, Th(NO3)4.3.67Me3PO, Th(NO3)4.3Me3PO, Th(NO3)4.2.67Me3PO,and Th(NO3)4.2.33Me3PO having been obtained. Thorium nitrate complexes display afascinating range of structure (Table 11.5).

11.6.2 Uranium(IV) Nitrate Complexes

Uranium(VI) nitrate complexes have been discussed in Section 11.5.4, but uranium formscomplexes in the +4 state that are generally similar to those of thorium.

Cs2U(NO3)6 + 2 Ph3PO → U(NO3)4(OPPh3)2 + 2 CsNO3

SPH SPH


Complexes of the Actinide(IV) Nitrates and Halides 181

Table 11.5 Thorium nitrate complexes

Formula Thorium species present Coord. No.

MTh(NO3)6 (M = Mg, Ca) [Th(NO3)6]2− 12Ph4P+ [Th(NO3)5(OPMe3)2]− [Th(NO3)5(OPMe3)2]− 12Th(NO3)4.5H2O [Th(NO3)4(H2O)3] 11Th(NO3)4.2.67Me3PO {[Th(NO3)3(Me3PO)4]+}2 [Th(NO3)6]2− 10, 12Th(NO3)4.2Ph3PO [Th(NO3)4(OPPh3)2] 10Th(NO3)4.5Me3PO [Th(NO3)2(OPMe3)5]2+ 9

Figure 11.6Structure of Th(NO3)4(H2O)3.

The reaction is carried out in a non-polar solvent like propanone; precipitated caesiumnitrate is filtered off and green (the most characteristic colour of UIV complexes) crystalsof the nitrate complex are obtained on concentrating the solution. U(NO3)4(OPPh3)2 has a10-coordinate structure (Figure 11.7) with phosphine oxide ligands trans to each other, andbidentate nitrates.

11.6.3 Complexes of the Actinide(IV) Halides

A large number of these have been synthesized, usually by reaction of the halides with theligand in a non-polar solvent like MeCN or acetone, which will form a labile complex suchas [UCl4(MeCN)4] that will undergo ready substitution by a stronger donor:

ThCl4 + 5 Me2SO → ThCl4(Me2SO)5 (in MeCN)

UCl4 + 2 (Me2N)3PO → UCl4[(Me2N)3PO]2 (in Me2CO)

Although sometimes direct reaction with a liquid ligand is possible:

UCl4 + 3 THF → UCl4(thf)3

ThCl4 + 3 Me3N → ThCl4(Me3N)3; UCl4 + 2 Me3N → UCl4(Me3N)2

Relatively few UI4 complexes have been made. They can often be synthesized by re-action in a solvent like MeCN and, whilst relatively stable thermally, undergo readyoxidation to uranyl complexes in (moist) air. The structures of many of these com-pounds have been determined, such as UCl4L2 [L = Ph3PO, (Me2N)3PO, Et3AsO,(Me2N)2PhPO] and UBr4L2 [L = Ph3PO, (Me2N)3PO] and UX4[(Me2N)2CO]2(X = Cl,Br, I). Both the 2:1 stoichiometry and trans-UX4L2 geometry are very common, but there are

SPH SPH



Figure 11.7Structure of U(NO3)4(Ph3PO)2.

exceptions. UCl4(Me2SO)3 is [UCl2(Me2SO)6] UCl6 and UCl4(Me3PO)6 is [UCl(Me3PO)6]Cl3; although no X-ray study has been carried out, UI4(Ph3AsO)2 is almost certainly[UI2(Ph3AsO)4] UI6.

Most structural reports concern uranium complexes. Many cases are known where com-plexes MX4Ln (n usually 2) exist for some or all the series Th–Pu; they are generallybelieved to have the same structure. UCl4(Ph3PO)2 is an exceptional compound in having acis geometry, confirmed by X-ray diffraction studies on crystals obtained on recrystalliza-tion from nitromethane (Figure 11.8). The phenyl rings in neighbouring Ph3PO moleculesface each other with a ring–ring separation of ∼3.52 A, an early recognized example ofπ–π stacking in a coordination compound. Similar compounds can be obtained for Th, Pa,Np, and Pu.

It was subsequently noted that crystals obtained immediately from the reaction mixturehad an IR spectrum different to that of authentic cis-UCl4(Ph3PO)2 but strongly resem-bling those of samples of trans-UBr4(Ph3PO)2. It seems likely that a less soluble trans-UCl4(Ph3PO)2 crystallizes first but on recrystallization or contact with the mother liquor itisomerizes to the thermodynamically more stable cis-UCl4(Ph3PO)2.

Figure 11.8Reproduced with permission from JCS Dalton, (1975) 1875. G. Bombieri et al. Copyright (1975)RSC.

SPH SPH


Thiocyanates 183

Table 11.6 Bond lengths in UX4L2 complexes

UCl4(tmu)2 UBr4(tmu)2 UI4(tmu)2 UCl4(hmpa)2 UBr4(hmpa)2

Average U–O (A) 2.209 2.197 2.185 2.23 2.18Average U–X (A) 2.62 2.78 3.01 2.615 2.781

Abbreviations: tmu = (Me2N)2C=O; hmpa = (Me2N)3P=O.

The only complete family of UX4 complexes (X = Cl, Br, I) that has been examinedcrystallograpically is that with tetramethylurea, UX4[(Me2N)2CO]2. Table 11.6 listsstructural data for these compounds and also for UX4[(Me2N)3PO]2(X = Cl, Br) {thoughthe compound UI4[(Me2N)3PO]2 has been made, its structure is not known}. Table 11.7shows the formulae of ThCl4 and UCl4 complexes with the same ligands. The ionic radiusof U4+ = 1.00 A and that of Th4+ = 1.05 A (values given for eight coordination, the pat-tern is similar for six coordination). Since the ions are of similar size, complexes usuallyhave similar stoichiometry (and indeed geometry). In a few cases (Me3N, Ph3PO, THF) theslightly greater size of thorium allows one more ligand to be attached to the metal ion.

Table 11.7 A comparison of thorium(IV) and uranium(IV) complexes isolated with the same ligand

Ligand Thorium Uranium

MeCN ThCl4(MeCN)4 UCl4(MeCN)4

Ph3PO ThCl4(Ph3PO)3; ThCl4(Ph3PO)2 UCl4(Ph3PO)2

(Me2N)3PO ThCl4[(Me2N)3PO]3; ThCl4[(Me2N)3PO]2 UCl4[(Me2N)3PO]2

(Me2N)2CO ThCl4[(Me2N)2CO]3 UCl4[(Me2N)2CO]2

Ph2SO ThCl4(Ph2SO)4 UCl4(Ph2SO)4; UCl4(Ph2SO)3

thf ThCl4(thf)3(H2O) UCl4(thf)3

Me3N ThCl4(Me3N)3 UCl4(Me3N)2

Me2N(CH2)2NMe2 ThCl4[Me2N(CH2)2NMe2]2 UCl4[Me2N(CH2)2NMe2]2

Me2P(CH2)2PMe2 ThCl4[Me2P(CH2)2PMe2]2 UCl4[Me2P(CH2)2PMe2]2

11.7 Thiocyanates

As the anhydrous actinide thiocyanates are not known {only hydrated [M(NCS)4(H2O)4]},thiocyanate complexes are prepared metathetically.

ThCl4(Ph3PO)2 + 4 KNCS + 2 Ph3PO → Th(NCS)4(Ph3PO)4 + 4 KCl (in Me2CO)

The precipitate of insoluble KCl is filtered off and the solution concentrated to obtainthe actinide complex. Several structures have been reported; Th(NCS)4(L)4 [L = Ph3PO,(Me2N)3PO]; U(NCS)4(L)4[L = Ph3PO, (Me2N)3PO, Me3PO] are all square antiprismatic;Th(NCS)4[(Me2N)2CO]4 is dodecahedral.

Some complexes are known where thiocyanate is the only ligand bound to uranium. Thegeometry of (Et4N)4 [U(NCS)8] is cubic, whilst Cs4 [U(NCS)8] is square antiprismatic.Compared with the d-block transition metals, there is not much evidence for directionalcharacter in bonding in lanthanide and actinide compounds. Because of this, the size andgeometry of the cation affects the packing arrangements in the lattice and energetically thisfactor must be more important than any crystal-field effects favouring a particular shape ofthe anion.

SPH SPH



11.8 Amides, Alkoxides and Thiolates

Compounds U(NR2)x and U(OR)x have been synthesized in oxidation states +3 to +6(though the hexavalent amides are not well characterized). A limited number of thiolatesU(SR)4 are also known. These largely molecular species occupy a borderline betweenclassical coordination chemistry and organometallic chemistry.

11.8.1 Amide Chemistry

Although they have been much less studied than the range of alkoxides in the +4, +5, and +6oxidation states, a number of amides of uranium(IV) have been made by ‘salt-elimination’reactions in solvents such as diethyl ether or THF, examples being:

UCl4 + 4 LiNEt2 → U(NEt2)4 + 4 LiCl

UCl4 + 4 LiNPh2 → U(NPh2)4 + 4 LiCl

UCl4 + excess of LiN(SiMe3)2 → UCl{N(SiMe3)2}3 + 3 LiCl

The steric demands of the ligands can be quantified in terms of the steric coordinationnumber, CNS [see J. Marcalo and A. Pires de Matos, Polyhedron, 1989, 8, 2431; CNS =the ratio of the solid angle comprising the Van der Waals’ spheres of the atoms of the ligandand that of the chloride ligand, when placed at a typical distance from the metal, values forthese amide groups are: NEt2− 1.67; NPh2

− 1.79; N(SiMe3)2− 2.17].

U(NPh2)4 is a monomer with a tetrahedral geometry, whilst in the solid state U(NEt2)4 isactually a dimer, [U2(NEt2)10], which contains five-coordinate uranium, with two bridgingamides (Figure 11.9). As N(SiMe3)2 is a much bulkier ligand with larger solid cone angle,it seems likely that there is not room for a fourth amide group round uranium, hence thenon-isolation of U{N(SiMe3)2}4, with MCl{N(SiMe3)2}3 (M = Th, U) being the mostsubstituted species obtained; the halogen can be replaced by other groups such as methylor tetrahydroborate. The compound U(NMe2)4 is harder to prepare. In the solid state it hasa chain trimeric structure; each uranium is 6 coordinate, NMe2 being less bulky than theother amides and has a smaller steric coordination number than even NEt2.

So far, the amides described have all been in the +4 state, but one rare three coordinate UIII

compound is U{N(SiMe3)2}3(pyramidal, like the lanthanide analogues). Unusual amides

UCl4 LiNEt2LiNEt2

[U(NEt2)4]2 Li[U(NEt2)5]Tl+

[U(NEt2)5]

ROH[U(OR)4]

(R = Me, Et)

Ph2NH

[U(NPh2)4]RSH

[U(SR)4] (R = Et, Pri, Bun)

U

NEt2

NEt2

NEt2

N

NEt2N

Et2N

Et2N U

Et Et

Et Et

Figure 11.9Synthesis, structure and reaction of U(NEt2)4.

SPH SPH


Amides, Alkoxides and Thiolates 185

have also been prepared with uranium in the +5 and +6 oxidation states. The first step inthe syntheses involved preparing anionic complexes with saturated coordination spheres:

UCl4 + 6 LiNMe2 → [Li(THF)]2[U(NMe2)6];

UCl4 + 5 LiNEt2 → [Li(THF)][U(NEt2)5]

The anions can then be oxidized to neutral molecular species using TlBPh4 or, better,AgI:

[U(NMe2)6]2− → [U(NMe2)6]− → [U(NMe2)6]; [U(NEt2)5]− → [U(NEt2)5]

[U(NEt2)5] (which can be obtained as dark red crystals) is a monomer in benzene solu-tion whilst [U(NMe2)6] is only known in solution. [U(NMe2)6] would be expected to beoctahedral and [U(NEt2)5] trigonal bipyramidal, but no structures are known. The mostversatile of all these compounds is U(NEt2)4.This is a low-melting (36 ◦C) and thermallystable substance (but like all alkoxides and alkylamides, immediately attacked by air or wa-ter). It can be distilled at ∼40 ◦C at 10−4 mmHg pressure and on account of this volatilitywas investigated for some time as a possible material for isotopic separation of uranium bygaseous diffusion. U(NEt2)4 is a useful starting material for the synthesis of other amides,alkoxides, and thiolates (Figure 11.9)

Its utility stems from the fact that alcohols and thiols are better proton donors (moreacidic) than amines, thus:

U(NEt2)4 + 4 EtOH → U(OEt)4 + 4 Et2NH

11.8.2 Alkoxides and Aryloxides

U(OR)n compounds (R = alkyl or aryl) exist in oxidation states between +3 and +6; inaddition, there are some uranyl alkoxides. An unusual feature is the number of thermallystable compounds in the otherwise unstable and rare +5 state.

A few uranyl alkoxides have been made, such as golden yellow [UO2(OCHPh2)2(thf)2],but more important are the large number of octahedral U(OR)6 compounds that can bemade, examples being U(OMe)6, U(OPri)6, U(OBut )6, U(OCF2CF3)6, and U(OCH2But )6

(U–O 2.001–2.002 A).

UF6 + 6 NaOMe → U(OMe)6 + 6 NaF

These compounds are volatile in vacuo, U(OCF2CF3)6 remarkably boiling at 25 ◦C at10 mmHg pressure. U(OMe)6, which sublimes at 30 ◦C at 10−5 mmHg pressure, was inves-tigated as a candidate for IR laser photochemistry leading to uranium isotopic enrichment.

Unlike the monomeric U(OR)6, U(OR)5 are usually associated; U(OPri)5 is dimeric withtwo alkoxide bridges giving six coordination. U(OEt)5 is the easiest compound to synthesizeand can be converted into others by alcohol exchange (Figure 11.10).

Uranium(IV) compounds tend to be non-volatile, highly associated solids, though thearyloxide U[O(2,6-But

2C6H3)]4 is made of monomeric tetrahedral molecules, doubtlessowing to its very bulky ligand. A number of UIII aryloxide species have been synthesized,examples including U[O(2,6-But

2C6H3)]3 being shown in Figure 11.11.Since four of the same aryloxide ligands can bind to uranium in the uranium(IV) com-

pound, a three coordinate U[O(2,6-But2C6H3)]3 is presumably coordinatively unsaturated,

and so forms a uranium-ring bond; coordinative saturation can also be achieved by formingan adduct with a Lewis base.

SPH SPH



U(NEt2)4 U(OEt)4

EtOH Br2 U(OEt)4Br

U(OEt)5

NaOEt

NaOEtUCl5

ROH (R, e.g., Pri)

U(OPri)5

Figure 11.10Synthesis of uranium alkoxides.

U

OR

ORO

CMe3

CMe3

ORO

RO

U Me3C

Me3C

[U{N(SiMe3)2}3]

ROH

2,6-But2C6H3OH

U(OR)3 U(OR)3.L

U(OR)4

O2

L(L = thf, CNBut, OPPh3)

KORUI4(MeCN)4

U(Me3Si)2N

(Me3Si)2N

N

CH2

SiMe2

SiMe3

exc. ROH

reflux

Figure 11.11Synthesis of uranium aryloxides.

11.9 Chemistry of Actinium

Little is known about the chemistry of actinium. It strongly resembles the lanthanides, espe-cially lanthanum, with identical reduction potentials (Ac3+ + 3e− → Ac; E = −2.62 V)and similar ionic radii (La3+ = 106 pm, Ac3+ = 111 pm). Its chemistry is dominated bythe (+3) oxidation state; as expected for a f 0 ion, its compounds are colourless, with no ab-sorption in the UV–visible region between 400 and 1000 nm. 227Ac is strongly radioactive(t1/2 = 21.77 y) and so are its decay products. Most of the work carried out has been on themicrogram scale, examining binary compounds such as the oxides and halides, and little isknown about its complexes. Actinium metal itself is a silvery solid, obtained by reduction ofthe oxide, fluoride, or chloride with Group I metals; it is oxidized rapidly in moist air. Likelanthanum, it forms an insoluble fluoride (coprecipitating quantitatively with lanthanum onthe tracer scale) and oxalate Ac2(C2O4)3.10H2O.

SPH SPH


Chemistry of Protactinium 187

Since the chemistry of actinium is confined to the Ac3+ ion, it can readily be separatedfrom thorium (and the lanthanides, for that matter) by processes like solvent extraction withthenoyltrifluoroacetone (TTFA) and by cation-exchange chromatography. The latter is anexcellent means of purification, as the Ac3+ ion is much more strongly bound by the resinthan its decay products.

11.10 Chemistry of Protactinium

Although its chemistry is not greatly studied at present, quite a lot of protactinium chemistryhas been reported. The main isotope is the alpha-emitter 231Pa, which has a long half-life(t1/2 = 3.28 × 104 y) so few problems arise, once appropriate precautions are taken withthe α-emission, the only other point of note being the ready hydrolysis of PaV in solution.Protactinium metal itself, formed by reduction (Ba) of PaF4, or thermal decomposition ofPaI5 on a tungsten filament, is a high-melting (1565 ◦C), dense (15.37 g cm−3), ductilesilvery solid, which readily reacts on heating with many non-metals.

The aqueous chemistry is dominated by the readily hydrolysed Pa5+ ion, which in theabsence of complexing ligands (e.g., fluoride) tends to precipitate as hydrated Pa2O5.nH2O.A PaO2

+ ion would be isoelectronic with the uranyl ion, UO22+, but there is no evidence

for it. However, solutions of PaV in fuming HNO3 yield crystals of PaO(NO3)3.x H2O(x =1–4), which may have Pa=O bonds; the sulfate PaO(HSO4)3 also exists. Reduction ofPaV with zinc or Cr2+ gives Pa4+(aq), only stable in strongly acidic solution, as, at higherpH, ions like Pa(OH)2

2+, PaO2+, and Pa(OH)3+ are believed to exist.

Many of the complexes of Pa that have been studied are halide complexes, falling intotwo types; anionic species with all-halide coordination, and neutral Lewis base adducts,usually of the chloride PaCl4.

Historically, the fluoride complexes are most important. Aristid Von Grosse (1934) usedK2PaF7 in his determination of the atomic mass of Pa in 1931. Colourless crystals of thisand other complexes MPaF6, M2PaF7, and M3PaF8 (M = K, Rb, Cs) can be obtained bychanging the stoichiometry, e.g.:

2 KF + PaF5 → K2PaF7 (in 17M HF, precipitate with propanone)

This complex has tricapped trigonal prismatic nine coordination of Pa, whilst MPaF6 hasdodecahedral 8 coordination and Na3PaF8 the very rare cubic 8 coordination. Other halidecomplexes can be made by appropriate methods:

PaX4 + 2 R4NX → (R4N)2PaX6 (X = Cl, Br; R = Me, Et; in MeCN)

PaCl4 + 2 CsCl → Cs2PaCl6 (in SOCl2/ICl)

PaI4 + 2 MePh3AsI → (MePh3As)2PaI6 (in MeCN)

Six-coordinate [PaX6]2− ions have been confirmed by X-ray diffraction for (Me4N)2PaX6

(X = Cl, Br) and Cs2PaCl6.Many adducts of the tetrahalides of ligands like phosphine oxides have been synthe-

sized by direct interaction of the tetrahalides with the ligands in solution in solvents likeacetonitrile or propanone. These have similar stoichiometries and structures to analogouscomplexes of Th, U, Np, and Pu, examples being trans-PaX4[(Me2N)3PO]2 (X = Cl, Br);cis-PaCl4(Ph3PO)2; PaCl4(MeCN)4; PaCl4(Me2SO)5 {possibly [PaCl3(Me2SO)5]+ Cl−}

SPH SPH



and PaCl4(Me2SO)3{possibly [PaCl2(Me2SO)6]2+ [PaCl6]2−}. In addition, a few adductsof the pentahalides, PaX5(Ph3PO)n(n = 1,2) exist.

In addition to these, a number of other complexes have been isolated, such as the β-diketonates [Pa(PhCOCHCOPh)4] and antiprismatic [Pa(MeCOCHCOMe)4], which, likethe TTFA complex [Pa(C4H3SCOCHCOCF3)4], are soluble in solvents like benzene, andsuitable for purification by solvent extraction. Pa forms a volatile borohydride [Pa(BH4)4].These compounds are typical of those formed by other early actinides, such as Th and U.

11.11 Chemistry of Neptunium

The metal itself is a dense (19.5 g cm−3) silvery solid, which readily undergoes oxidationin air. Chemical study uses one isotope, the long-lived 237Np (t1/2 = 2.14 × 106 y). Thechemistry of neptunium shows interesting points of comparison between U and Pu, whichflank it (and like them it shows a wide range of oxidation states). Thus, whilst the (+6)oxidation state is found in the [NpO2]2+ ion as well as in the halide NpF6 (note that thereis no NpCl6, unlike U), the (+6) state is less stable than in the case of uranium, thoughit is more stable than PuVI. However, the (+5) state is more stable for neptunium thanfor uranium, with the [NpO2]+ ion showing no signs of disproportionation, though easilyundergoing reduction by Fe2+ to Np4+.

Np4+ is in many ways the most important oxidation state. It is formed by reduction of thehigher oxidation states, and by aerial oxidation of Np3+. Strong oxidizing agents like Ce4+

oxidize it back to [NpO2]2+, whilst electrolytic reduction of Np4+ affords Np3+, which isstable in the absence of air (unlike U).

As neptunium has one more outer-shell electron than uranium, it has the possibility ofa (+7) oxidation state, a possibility realized in alkaline solution, when ozone will oxidizeNpVI to NpVII, an oxidation also achieved by XeO3 or IO4

− at higher temperatures. Thepotential for this is estimated as −1.24V (1M alkali). The relevant standard potentials forneptunium (1M acid) are:

Np3+ + 3 e− → Np E = −1.79 VNp4+ + e− → Np3+ E = −0.15 VNpO2

+ + 4 H+ + e− → Np4+ + 2 H2O E = +0.67 VNpO2

2+ + e− → NpO2+ E = +1.24 V

11.11.1 Complexes of Neptunium

A significant amount of structural information has been gathered on neptunium complexes.For the main, they resemble corresponding U and Pu complexes. However, an EXAFSstudy of alkaline solutions of NpVII found evidence for a trans dioxo ion of the type[NpO2(OH)4(OH2)]1−.

In the VI state, the 8 coordinate Na[NpO2(OAc)3], [NpO2(NO3)2(H2O)2].4H2O,K4[NpO2(CO3)3], and Na4[NpO2(O2)3].9H2O [where (O2)3 indicates three peroxide di-anion ligands] are isostructural with the U, Pu, and Am analogues. They demonstrate acontraction of ∼0.01 A in the M=O distance per unit increase in atomic number. Eightcoordination is also found in [NpO2(NO3)2(bipy)]. The aqua ion appears to be seven

SPH SPH


Chemistry of Plutonium 189

coordinate [NpO2(H2O)5]2+. Like the uranyl ion, the neptunyl ion complexes with ex-panded porphyrins like hexaphyrin, amethyrin, pentaphyrin, and alaskaphyrin.

In the NpV state, structural comparison of Ba[NpVO2(OAc)3] with Na[NpVIO2(OAc)3]indicates a lengthening of 0.14 A in the neptunium–oxygen bond length on going fromNpVI to NpV, consistent with an electron added to an antibonding orbital. As with theV state, the aqua ion is believed to be seven coordinate [NpO2(H2O)5]+; in the related[NpO2(urea)5](NO3) the neptunium atom has a typical pentagonal-bipyramidal environ-ment with five oxygen atoms in the equatorial plane. A similar geometry is found in theacetamide complex of neptunium V nitrate, [(NpO2)(NO3)(CH3CONH2)2], where the pen-tagonal bipyramidal coordination of Np is completed (in the solid state) using oxygens ofneighbouring NpO2 groups; similarly in NpO2ClO4.4H2O, the neptunyl(V) ion has four wa-ters and a distant neptunyl oxygen occupying the equatorial positions. Eight and six coordi-nation are, respectively, found in the crown ether complex, [NpO2(18-crown-6)] ClO4, and in[NpO2(OPPh3)4]ClO4. Carbonate complexes have been investigated as they could representa means of leaching transactinides from underground deposits. EXAFS studies of NpV car-bonate complexes indicate the existence of [NpO2(H2O)3(CO3)]−, [NpO2(H2O)2(CO3)2]3−

and [NpO2(CO3)3]5−; similar NpVI species like [NpO2(CO3)3]4− are indicated.Among neptunium(VI) complexes, [Np(S2CNEt2)4] and [Np(acac)4] have dodecahedral

and antiprismatic coordination of neptunium. A reminder of the small differences in energybetween different geometries is that [Me4N]4 [Np(NCS)8] has tetragonal antiprismaticcoordination, whilst [Et4N]4 [Np(NCS)8] has cubic coordination of Np. Neptunium(IV) isalso eight coordinate in [Np(urea)8]SiW12O40·2Urea·H2O.

As with Th, Pa, U, and Pu, a variety of neutral complexes are formed between the tetra-halides (and nitrate) and ligands like phosphine oxides, such as trans-[NpX4{(Me2N)3PO}2](X = Cl, Br); cis-[NpCl4(Ph3PO)2]; NpCl4(Me2SO)3 (possibly [NpCl2(Me2SO)6]2+

[NpCl6]2−); [Np(NCS)4(R3PO)2] (R = Ph, Me, Me2N), and [Np(NO3)4(Me2SO)3].Only a few complexes of NpIII have been characterized, notably the dithiocarbamate Et4N

[Np(S2CNEt2)4], which has distorted dodecahedral coordination like the Pu and lanthanideanalogues. [NpI3(thf)4] and the silylamide [Np{N(SiMe3)2}3] are analogous to those ofU and the lanthanides, the former potentially being a useful starting material. The amidereacts with a bulky phenol:

[Np{N(SiMe3)2}3] + 3 ArOH → [Np(OAr)3] + 3 HN(SiMe3)2

(Ar = 2,6-But2C6H3)

11.12 Chemistry of Plutonium

Plutonium presents particular problems in its study. One reason is that, since 239Pu is astrong α-emitter (t1/2 = 24,100 years) and also tends to accumulate in bone and liver, it isa severe radiological poison and must be handled with extreme care. A further problem isthat the accidental formation of a critical mass must be avoided.

11.12.1 Aqueous Chemistry

Because of the complicated redox chemistry, not all the oxidation states +3 to +7 are ob-served under all pH conditions. Blue Pu3+(aq) resembles the corresponding Ln3+ ions whilstthe brown Pu4+ ion requires strongly acidic conditions (6M) to prevent oligomerization

SPH SPH



and disproportionation. The purple-pink PuO2+ ion readily disproportionates, whilst

the orange-yellow PuO22+ ion resembles the uranyl ion, but is less stable, tending to

be reduced, not least by its own α-decay. Blue PuVII, possibly PuO53−(aq) or some

oxo/hydroxy species like [PuO2(OH)4(OH2)]3−, exists only at very high pH; it is formedby ozonolysis of PuVI, and some PuVIII may be generated.

11.12.2 The Stability of the Oxidation States of Plutonium

In comparison with uranium, the (+3) state has become much more stable (Table 9.5),witness the standard reduction potential of +0.98 V for Pu4+ + e → Pu3+, the comparativevalue for uranium being −0.63V. This means that quite strong oxidizing agents such asmanganate(VII) are needed to effect this oxidation in the case of plutonium.

The relevant standard potentials for plutonium are:

Pu3+ + 3 e− → Pu E = −2.00 VPu4+ + e− → Pu3+ E = +0.98 VPuO2

+ + 4 H+ + e− → Pu4+ + 2 H2O E = +1.04 VPuO2

2+ + e− → PuO2+ E = +0.94 V

The PuIII–PuIV and the PuV–PuVI couples are both reversible, but not the PuIV–PuV, as the latterinvolves the making and breaking of Pu=O bonds and significant changes in geometry (cf.uranium). Reactions involving the making and breaking of Pu=O bonds are also kineticallyslow, so that it is possible for ions in all four oxidation states between (+3) and (+6) to coexistin aqueous solution under certain conditions. The even spacing of the potentials linking thefour oxidation states means that disproportionation and reproportionation reactions arefeasible. For example, a possible route for disproportion of Pu4+ in aqueous solution canbe written:

2 Pu4+ + 2 H2O → Pu3+ + PuO2+ + 4 H+

This is composed of two redox processes:

Pu4+ + e− → Pu3+ and Pu4+ + 2 H2O → PuO2+ + 4 H+ + e−

Because the potentials are of roughly equal magnitude (and of opposite sign in this case),the free energy change is small (<20 kJmol−1). The following process:

2 PuO2+ + 4 H+ → PuO2

2+ + Pu4+ + 2 H2O

is also favoured; overall, the two can be combined as:

3 Pu4+ + 2 H2O → 2 Pu3+ + PuO22+ + 4 H+.

Another feasible reproportionation reaction is of course:

PuO2+ + Pu3+ + 4 H+ → 2 Pu4+ + 2 H2O

A further problem is that compounds in the +5 and +6 oxidation states tend to bereduced (autoradiolysis), as 239Pu is a strong α-emitter (1 mg 239Pu emits over a millionα-particles a second) decomposing water molecules into •H, •OH and •O radicals which,in turn, participate in redox reactions. In acidic solution, decomposition of the solvent toH2O2 and the acid can occur. In nitric acid, for example, both HNO2 and nitrogen oxides areformed, so that, starting with plutonium(VI), ions in the lower oxidation states are generated

SPH SPH



Figure 11.12(a) Disproportionation of PuV in 0.1M HNO3 + 0.2 M NaNO3. (b) Self-reduction of PuVI due toits own α-emission, in 5M NaNO3 (after P.I. Artiukhin, V.I. Medicedovskii, and A.D. Gel’man,Radiokhimiya, 1959, 1 131; Zh. Neorg. Khim., 1959, 4, 1324) (permission applied for reproduction).

by both the reduction and disproportionation/reproportionation reactions described above.Figure 11.12 shows the effects of these processes. In Figure 11.12(a), the disproportionationof PuV into all the (+3) to (+6) states can be seen, whilst in Figure 11.12(b), over a longertime scale, the self-reduction of PuVI to PuIV is observed.

In perchloric acid, the (+4) state is less abundant, and the (+3) state more abundant, areflection of the higher oxidizing power of nitric acid.

11.12.3 Coordination Chemistry of Plutonium

In its coordination chemistry, plutonium(VI) resembles uranium, where corresponding com-plexes exist. Thus crystallization of solutions of PuVI in concentrated nitric acid affordsred-brown to purple crystals of the hexahydrate PuO2(NO3)2.6H2O, which is isostruc-tural with the U analogue, and contains 8-coordinate [PuO2(NO3)2(H2O)2] molecules. AnX-ray absorption spectroscopy study of tributyl phosphate complexes of [AnO2(NO3)2](An = U, Np, Pu) in solution, probably present as [AnO2(NO3)2(tbp)2], has indicated ma-jor changes in the actinide’s coordination sphere on reduction to the AnIV state, but nosignificant changes across the series UO2

2+, NpO22+, PuO2

2+. An actinide contraction of

SPH SPH



about 0.02 A between successive actinides was detected (M = O changes overall from 1.79to 1.75 A on passing from U to Pu).

Adding acetate ions to a PuVI solution, in the presence of the appropriate Group I metal(M), results in precipitation of M [PuO2(OAc)3], also 8 coordinate, with bidentate acetates,isostructural with the U, Np, and Am analogues. Other examples of polyhedra resemblingthe corresponding uranyl complexes are to be found in the [PuO2Cl4]2−(trans-octahedral)and [PuO2F5]3− (pentagonal bipyramidal) ions. Plutonium(VI) carbonate complexes areimportant in the preparation of ceramic materials. The PuO2

2+ ion is precipitated fromaqueous solution by carbonate ion, but this is soluble in excess of carbonate.

PuO22+(aq) → PuO2CO3(s) → [PuO2(CO3)2]2−(aq) and [PuO2(CO3)3]4−(aq)

In the latter ion, Pu has (2 + 6) coordination {X-ray diffraction of [C(NH2)4][PuO2(CO3)3]}. The ammonium salt undergoes two-stage thermal decomposition, via (theisolable) PuO2CO3:

(NH4)4 [PuO2(CO3)3] → PuO2CO3 + 4 NH3 + 2 H2O + 2 CO2 (at 110–150 ◦C)

PuO2CO3 → PuO2 + 1/2 O2 + CO2 (at 250−300 ◦C)

Few Pu(+5) compounds have been characterized in the solid state.In the (+4) oxidation state, plutonium forms many halide complexes, both anionic and

neutral, contrasting with the absence of PuCl4 and PuBr4 (though no iodides are known).Pu4+(aq) reacts with conc. HX forming [PaX6]2− ions (X = Cl, Br) isolable as salts suchas (Et4N)2PuBr6 and Cs2PuCl6, both probably containing octahedrally coordinated Pu, anduseful as starting materials for making PuIV halide complexes. Various fluoride complexescan be made, often relying on the instability of PuF6:

LiF + PuF6 → LiPuF5 (simplified equation) (at 300 ◦C)

LiPuF5 → Li4PuF8 (at 400 ◦C)

CsF + PuF6 → Cs2PuF6 (at 300 ◦C)

NH4F + PuF6 → NH4PuF5 + (NH4)2PuF6 (at 70–100 ◦C)

High coordination numbers are usual in fluoride complexes; (NH4)4PuF8 has nine-coordinate Pu.

Neutral halide complexes are usually got starting from [PuX6]2− salts (X = Cl, Br),though bromides can also be made by bromine oxidation of PuIII species.

Cs2PuCl6 + 2 Ph3PO → cis-PuCl4(Ph3PO)2 + 2 CsCl (in MeCN)

PuBr3 + 1/2 Br2 + 2 Ph3PO → trans-PuBr4(Ph3PO)2 (in MeCN)

A wide range of these complexes, also involving nitrate and thiocyanate ligands, has beenmade:

Cs2Pu(NO3)6 + 2 Ph3PO → Pu(NO3)4(Ph3O)2 + 2 CsNO3

PuCl4(R3PO)2 + 2 R3PO + 4 KNCS → Pu(NCS)4(R3PO)4 + 4 KCl

(R, e.g., Ph, Me2N)

Others include trans-PuX4[(Me2N)3PO]2 (X = Cl, Br); PuCl4(Ph2SO)n(n = 3, 4);PuCl4(Me2SO)n (n = 3, 7); PuCl4(Me3PO)6 {probably [PuCl(Me3PO)6]3+ (Cl−)3}. Struc-turally, they resemble the corresponding complexes of Th, Pa, U, and Np. The tridentate

SPH SPH



ligand 2,6-[Ph2P(O)CH2)]2C5H3NO combines two phosphine oxide and one N-oxide func-tional groups; it forms a 2:1 PuIV complex which has the same structure, [PuL2(NO3)2](NO3)2, in both the solid state and solution. The 2:1 complex of the bidentate ligand2-[Ph2P(O)CH2]C5H4NO with plutonium nitrate also has an ionic structure [PuL2(NO3)3]2

[Pu(NO3)6]. As is the case with uranium, nitrate complexes are important in the separationchemistry of plutonium. Solutions of Pu4+(aq) in conc. HNO3 deposit dark green crystalsof Pu(NO3)4.5H2O, which contain 11-coordinate [Pu(NO3)4(H2O)3] molecules (as is thecase with Th). [Pu(NO3)6]2− ions can be fished out of very concentrated (10–14M) nitricacid solutions as salts R2Pu(NO3)6(R = Cs, Et4N). [Pu(NO3)6]2− is strongly adsorbed byanion-exchange resins, and this is made use of in the purification of Pu commercially.

Plutonium carbonate complexes are important as they contribute to an understand-ing of what happens to plutonium in the environment, Pu(+4) being the most sta-ble oxidation state under normal conditions. [Pu(CO3)5]6− has been identified byEXAFS as the PuIV species in solution at high carbonate concentration; the struc-ture of crystalline [Na6Pu(CO3)5]2.Na2CO3.33H2O has also been determined; it fea-tures 10-coordinate plutonium. The sulfate complex K4Pu(SO4)4.2H2O contains dimeric[(SO4)3Pu(µ-SO4)2Pu(SO4)3]8− anions with nine-coordinate Pu.

In many aspects of its coordination chemistry, PuIV compounds resemble their ura-nium analogues; thus it forms eight-coordinate complexes with diketonate ligands, like[Pu(acac)4], and similar complexes with 8-hydroxyquinolinate and tropolonate. There is alimited alkoxide chemistry, [Pu(OBut )4] being volatile at 112 ◦C (0.05 mmHg pressure) andvery likely being a monomer. The borohydride [Pu(BH4)4] is a blue-black volatile liquid(mp 15 ◦C) and has a 12 coordinate molecular structure, with tridentate borohydride ligands,like the neptunium analogue, but unlike 14 coordinate [U(BH4)4].

Little coordination chemistry is as yet known in the (+3) state, but these compoundsare interesting. Reaction of Pu with iodine in THF affords off-white [PuI3(thf)4], a use-ful starting material; [PuI3(dmso)4] and [PuI3(py)4] have also been reported. The σ -alkyl[Pu{CH(SiMe3)2}3] and the silylamide [Pu{N(SiMe3)2}3] are analogous to those of U andthe lanthanides, the latter potentially being a useful starting material (though neither hasbeen completely characterized). The amide reacts with a bulky phenol forming the arylox-ide [Pu(OAr)3] (Ar = 2,6-But

2C6H3), which is probably three coordinate. Plutonium formsinsoluble oxalates Pu2(C2O4)3.x H2O (x = 10, 11) similar to those of the lanthanides.

Reaction of plutonium metal with triflic acid affords blue Pu(CF3SO3)3.9H2O, isostruc-tural with the lanthanide triflates and which contains tricapped trigonal prismatic[Pu(H2O)9]3+ ions; similarly, Pu reacts with AgPF6 (or TlPF6) in MeCN suspension, dis-solving to form [Pu(MeCN)9] (PF6)3, in which the coordination polyhedron is rather moredistorted.

11.12.4 Plutonium in the Environment

Actinides are potentially present in the environment from a number of sources, not justthe ‘natural’ thorium and uranium. Nuclear power plants and uses in nuclear weapons areobvious areas for concern, raised in recent times by possibilities that terrorist groups couldgain control of such weapons, possibly ‘dirty’ bombs. Plutonium is an obvious sourceof concern. Any plutonium in the environment is believed to be largely present as ratherinsoluble PuIV species. PuIV ions are hydrolysed easily, unless in very acidic solution, forminglight green colloidal species, which age with time, their solubility decreasing. The possibilityof colloid-facilitated transport of plutonium occurring was raised by the observation of

SPH SPH



migration of plutonium in ground water by rather more than a mile from the location ofunderground weapons testing in Nevada, USA, over a period greater than 20 years. It istherefore necessary to consider the interaction of plutonium, not just with inorganic ionslike carbonate (see Section 11.12.3), phosphate, silicate and sulfate, but with natural organicsubstances like humic acid.

Disposal of nuclear waste materials – especially the high-level waste from the cores ofreactors and nuclear weapons – so that they do not escape into the environment is a pressingproblem. Currently the preferred solution in the USA and most of Europe is immobilizationin a suitable geological repository. The waste materials are evaporated if necessary, thenheated with silica and borax to form a borosilicate glass, encased in an inert (e.g., lead)material, then buried in a stable zone, such as rock salt or clay. The conditions surroundingsuch sites are important. Yucca Mountain, Nevada, is the leading candidate among possibleAmerican sites. Solubility studies have shown that Np is more than 1000 times more solublein the Yucca Mountain waters than is Pu, because under the ambient conditions Pu tendsto adopt the (+4) state, associated with insolubility, whilst Np adopts the (+5) state, wherecompounds are much more soluble. However, at the WIPP site in New Mexico, whichis built on a deep salt formation, the very salty brines favour formation of soluble Pu(+6)complexes such as [PuO2Cl4]2−, which will present much more of a problem unless reducingconditions can be generated.

As a potential environmental hazard, plutonium is particularly toxic because of the re-markable similarity of PuIV and FeIII – which means that Pu can be taken up in the biologicaliron transport and storage system of mammals. This similarity is being used in a biomimeticapproach to Pu-specific ligands for both in-vivo actinide decorporation and nuclear wasteremediation agents. It has also been suggested that plutonium could be solubilized by mi-crobial siderophores. These are ligands used by bacteria to bind Fe3+ (an ion which, likePu4+, is essentially ‘insoluble’ at near-neutral pH) and carry it into cells. In fact, suchsiderophores bind plutonium strongly (log β = 30.8 for the desferrioxamine B complex),only taking it up as PuIV with plutonium in other oxidation states being oxidized or re-duced appropriately. They do not solubilize plutonium rapidly, however; they are, in fact,less effective in solubilizing plutonium than are simple complexing agents like EDTA andcitrate, leading to the suggestion that the siderophores passivate the surface of the pluto-nium hydroxide. (The overall formation constant for [Pu-EDTA] in acidic solution is logβ = 26.44; the strength of the complexation means that the possibility of EDTA, widelyused in former nuclear weapons programmes, solubilizing and transporting plutonium is apresent concern).

Unlike the exclusively 6-coordinate iron(III) siderophore complexes, higher coordina-tion numbers are possible with plutonium; the complex of PuIV with desferrioxamine E(DFE; a hexadentate iron-binding siderophore ligand) has shown it to contain 9-coordinate[Pu(dfe)(H2O)3]+ ions with a tricapped trigonal prismatic geometry (Figure 11.13).

The microbe Mycena flavescens has been found to take up Pu (but not U) in the formof a siderophore complex though at a much slower rate than it takes up iron; the Fe andPu complexes inhibit each other, indicating competition for the same binding site on themicrobe. It has similarly been found that pyoverdin, the main siderophore in iron-gatheringcapacity produced by Pseudomonas aeruginosa, will also transport plutonium across cellmembranes. It has been suggested that using phytosiderophores such as desferrioxamine andmugineic acids could solubilize actinides and promote their uptake by plants and consequentremoval from the soil.

SPH SPH


Chemistry of Americium and Subsequent Actinides 195

OH2N

N

O O

O

N

OO

ON

N

O

O

O

N

H

H

HPu

H2O

H2O

+

Figure 11.13Structure of a Puiv complex of a siderophore.

11.13 Chemistry of Americium and Subsequent Actinides

Two isotopes, 241Am (t1/2 = 433 y), a decay product of 241Pu, and 243Am (t1/2 = 7380 y),have half-lives suitable for their use in chemical reactions. The metal itself is a silvery,ductile and malleable solid, which is obtained by metallothermic (Ba, Li) reduction of thetrifluoride; alternatively by heating the dioxide with lanthanum and using the difference inboiling point between La (bp 3457 ◦C) and americium (2607 ◦C) to displace the equilibriumto the right as americium distils.

3 AmO2 + 4 La � 2 La2O3 + 3 Am

Americium undergoes slow oxidation in air and dissolves in dil. HCl, as expected from itsfavourable potential [E(Am/Am3+) = −2.07 V].

11.13.1 Potentials

The relevant standard potentials for americium (1M acid) are:

Am3+ + 3e− → Am E = −2.07 VAm4+ + e− → Am3+ E = +2.3 VAmO2

+ + 4H+ + e− → Am4+ + 2 H2O E = +0.82 VAmO2

2+ + e− → AmO+2 E = +1.60 V

The tendency for the (+3) state to become more stable in the sequence U < Np < Pu con-tinues with americium (and dihalides, AmX2, make their appearance for the first time). Thepink Am3+ ion is thus the most important species in aqueous solution, AmIV being unsta-ble in the absence of complexing agents. Hypochlorite oxidizes Am3+ in alkaline solutionto what may be Am(OH)4, soluble in NH4F solution, very likely as the fluoride complex[AmF8]4−. The (+5) state is accessible, again through oxidation in alkaline solution (e.g.,with O3 or peroxydisulfate, S2O8

2−), as the AmO2+ ion, though this tends to dispropor-

tionate to AmIII and AmV; the (+6) state, in the form of the AmO22+ ion, is obtained by

oxidation (Ag2+ or S2O82−) of lower oxidation states in acid solution.

SPH SPH



The tendency noted for plutonium for disproportionation occurs here, as both AmIV andAmV tend to disproportionate in solution:

2 AmIV → AmIII + AmV

Am4+ + AmO2+ → AmO2

2+ + Am3+

3 AmO2+ + 4 H+ → 2 AmO2

2+ + Am3+ + 2 H2O

There is also again the tendency to autoradiolysis, the reduction of high oxidation statesthrough the effects of the α-radiation emitted.

Relatively few complexes of americium have been characterized; those that have tendto resemble the corresponding compounds of the three previous metals. Many are halidecomplexes, such as (NH4)4 [AmF8], which resembles the U analogue, Cs2NaAmCl6, and(Ph3PH)3AmX6 (X = Cl, Br).

The AmO2+ and AmO2

2+ ions form well-defined complexes; thus in HCl solution, whereit likely forms [AmO2Cl4]3− and [AmO2Cl4]2−, the symmetric Am=O stretching vibrationscan be detected at 730 (AmO2

+) and 796 cm−1 (AmO2+), respectively. Na[AmO2(OAc)3]

has eight coordination with bidentate acetates, just like the U–Pu analogues, the same coordi-nation number also being found in AmO2F2, M [AmO2F2] (M = Rb, K), and M [AmO2CO3](M = Rb, K, Cs). Six coordination occurs in Cs2[AmO2Cl4] and in NH4[AmO2PO4].

A significant chemistry occurs in the (+3) state, similar to that of Ln3+ ions, withstudies often made in the context of separating Am from lanthanide fission products.Amide ligands like N, N, N ′, N ′-tetraethylmalonamide (TEMA) have been investigatedas possible extractants. Solution studies are assisted by some of the newer spectro-scopic techniques; thus EXAFS (Extended X-ray Absorption Fine Structure) studies onsolutions of [Am(TEMA)2(NO3)3] indicate a similar geometry to [Nd(TEMA)2(NO3)3].Carboxylate-derived calix[4]arenes show high selectivity for Am3+, whilst complexationof Am3+ by crown ethers and diazacrown ethers has also been studied. The resemblanceextends to complexes isolated. The sulfate Am2(SO4)3.8H2O is isomorphous with the8-coordinate lanthanide analogies (Pr–Sm) and insoluble oxalates Am2(C2O4)3.x H2O(x = 7, 11) have been characterized. The resemblance extends to diketonate complexessuch as [{Am(Me3CCOCHCOCMe3)3}2], a dimer with seven-coordinate Am, like the Pranalogue. The adduct [Am(CF3CCOCHCOCCF3)3{(BuO)3PO}2] is volatile at 175 ◦C andpotentially could be used in separations. In the iodate complex K3Am3(IO3)12·HIO3, the[AmO8] polyhedra are made of eight [IO3] oxygen atoms in a distorted bicapped trigonalprismatic array. There is additionally one very long Am–O contact to complete a distortedtricapped trigonal prismatic Am coordination sphere.

11.14 Chemistry of the Later Actinides

The succeeding actinides (Cm, Bk, Cf, Es, Fm, Md, No, Lr) mark the point where the listof isolated compounds tends to involve binary compounds (oxides, halides and halide com-plexes, chalcogenides, and pnictides) rather than complexes. Those studies of complexesthat have been made are usually carried out in solution and, from Fm, onwards, have beentracer studies.

Fermium coprecipitates with lanthanide fluorides and hydroxides, showing it to formlanthanide-like Fm3+ ions. These elute from cation-exchange resins slightly before Es3+,whilst its chloride and thiocyanate complexes are eluted from anionic exchange resins just

SPH SPH


Chemistry of the Later Actinides 197

after Es3+. This shows that Fm3+ forms slightly stronger complexes than does Es3+, asexpected purely on electrostatic considerations for a slightly smaller ion. In the case ofnobelium, slightly different results have been obtained. Nobelium ions coprecipitate withBaF2, rather than with LaF3, suggesting that they are present as No2+ ions, but, after addingCe4+ oxidant to the initial solution, nobelium precipitated with LaF3, indicating that thecerium had oxidized the nobelium to No3+ ions. Complexing studies using Cl− ions alsoindicated alkaline-earth-like behaviour.

Question 11.1 Use the reduction potentials to show why the UO2+(aq) tends to dispropor-

tionate.

UO2+(aq) + 4 H+(aq) + e− → U4+(aq) + 2 H2O(l) E = +0.62 V

UO22+(aq) + e− → UO2

+(aq) E = +0.06 V

Answer 11.1 Combining the potentials shows that for

2 UO2+ (aq) + 4 H+(aq) → U4+(aq) + UO2

2+(aq) + 2 H2O(l) E = +0.56 V

Since E is positive, the reaction is energetically feasible; although this only predicts that theUO+

2 (aq) ion is thermodynamically unstable with respect to disproportionation and saysnothing about kinetic stability, the fact is that the uranium(V) aqua ion is very short-lived.

Question 11.2 Write the expression for K1 and K2 for complex formation between Th4+

and F− ions.Answer 11.2

K1 = [ThF3+(aq)]/[Th4+(aq)][F−(aq)]; K2 = [ThF22+(aq)]/[ThF3+(aq)][F−(aq)]

Question 11.3 The highest occupied orbitals in the uranyl ion have the electronic ar-rangement σu

2 σg2 πu

4 πg4. By referring to Figure 11.2, suggest the electronic arrangements

similarly of UO2+ and PuO2

2+.Answer 11.3 UO2

+ is σu2 σg

2 πu4 πg

4 (δu, φu)1 and PuO22+ is σu

2 σg2 πu

4 πg4

(δu, φu)2 (The relative positions of δu and φu are uncertain).

Question 11.4 Suggest reasons why actinides from Cm onwards do not appear to formsimilar ions like CmO2

2+.Answer 11.4 Heavier actinides such as Cm do not display high oxidation states as thef orbitals appear to be more contracted and their electrons are not available for bonding,(and could be required for π bonding, for example). A further factor is that the additionalelectrons would need to be placed in antibonding orbitals such as σu

∗ and πu∗, which would

destabilize such an ion.

Question 11.5 Suggest coordination numbers for each of the following complexes.(a) [UO2F5]3−; (b) [UO2Br4]2−; (c) [UO2(NO3)3]−; (d) [UO2(NCS)5]3−; (e) [UO2

(CH3COO)3]−; (f) [UO2(NO3)2(H2O)2]; (g) [UO2(NO3)2(Ph3PO)2]; (h) [UO2(CH3COO)2

(Ph3PO)] .Answer 11.5 (a) 7; (b) 6; (c) 8 (nitrate is bidentate) (d) 7; (e) 8 (acetate is bidentate) (f) 8;(g) 8; (h) 7.

SPH SPH



Question 11.6 Explain why you can get 2 + 6 coordination in [UO2(OH2)2(O2NO)2],[UO2(O2NO)3]−, etc. rather than the 2 + 5 coordination in the uranyl aqua ion.Answer 11.6 Although bidentate in most complexes, nitrate groups take up little space(they are said to have a small ‘bite angle’).

Question 11.7 Read section 11.5.5 and explain why Pu4+ can be reduced by Fe2+, butUO2

2+ can’t be so reduced.Answer 11.7 Fe3+(aq) has a more negative reduction potential than Pu4+(aq), makingFe2+ a better reducing agent, one that will reduce Pu4(aq). Thus for:

Pu4+(aq) + Fe2+(aq) → Pu3+(aq) + Fe3+(aq) E = +0.23 V

and the process is energetically feasible. However, for:

UO22+(aq) + 4H+(aq) + 2Fe2+(aq) → U4+(aq) + 2H2O (l) + 2Fe3+(aq)

E = −0.50 V

and the reduction is not energetically feasible under these conditions.

Question 11.8 The infrared spectrum of Th(NO3)4.4 Me3PO displays bands due to biden-tate nitrate and ionic nitrate groups. Suggest a possible structure affording a reasonablecoordination number.Answer 11.8 Phosphine oxide ligands and bidentate nitrate groups take up differentamounts of space round thorium. Assuming that all the phosphine oxide groups are boundto thorium, reference to Table 11.5 suggests that a coordination number of 10 is likely;four Me3PO groups bound would leave room for three bidentate nitrates. The suggestedstructure is [Th(NO3)3(Me3PO)4]+NO3

−.

Question 11.9 (Refer to section 11.6.2 in connection with this) Hexamethylphospho-ramide (HMPA), OP(NMe2)3, is a good σ -donor, behaving like a typical monodentatephosphine oxide ligand such as Ph3PO. A number of complexes of HMPA with uranium(IV)nitrate have been prepared. Green crystals of U(NO3)4(hmpa)2 are made thus:

Cs2(NO3)6 + 2 HMPA (acetone) → 2 CsNO3 + U(NO3)4(hmpa)2

The IR spectrum of this compound shows only one type of nitrate group; all are coordinated.

Question A How are the nitrates likely to be bound in U(NO3)4(hmpa)2? Suggest a coor-dination number and coordination geometry for uranium.Answer A By analogy with U(NO3)4(OPPh3)2, a 10-coordinate compound with a transgeometry and bidentate nitrates is probable (this is known to be the case).

Using excess of HMPA, a green compound U(NO3)4(hmpa)4 is obtained. Its IR spectrumshows two different modes of nitrate coordination, and no ionic nitrate.

Question B Suggest a credible structure for the compound U(NO3)4(hmpa)4, indicatingthe coordination number of uranium.Answer B Various structures are possible. It is unlikely that all nitrates are bidentate, asthis product would be 12 coordinate. If one nitrate is monodentate, the compound wouldbe 11 coordinate U(O2NO)3(ONO2)(hmpa)4.

SPH SPH


Chemistry of the Later Actinides 199

If U(NO3)4(hmpa)4 is dissolved in non-polar solvents such as CH3CN, a non-conductingsolution is obtained, but with a simpler IR spectrum possibly indicating just one type ofnitrate coordination.

Question C Suggest a structure for the compound present in solution.Answer C One alternative is an 11-coordinate complex U(NO3)4(hmpa)3 formed bydissociation of one hmpa ligand.

When U(NO3)4(hmpa)4 is treated with NaBPh4, a crystalline solid is obtained which hasthe analysis U(NO3)3(hmpa)4BPh4. This is a 1:1 electrolyte in solution. Its IR spectrumshows only one type of nitrate group, and no ionic nitrate.

Question D Suggest a credible structure for this compound, indicating the coordinationnumber of uranium.Answer D Assuming all nitrates are bidentate and that the BPh−

4 group remains ionic, astructure [U(NO3)3(hmpa)4]+(BPh4)− would contain 10-coordinate uranium.

Question 11.10 Comment on the patterns in bond length in Table 11.6.Answer 11.10 The U–X distances increase with increasing radius of halogen, as expected.There appears to be a slight decrease in U–O distance as the σ -donor power of the halogendecreases, possibly because this means that the ligands are competing for metal orbitals.

Question 11.11 Complete these equations:

UCl{N(SiMe3)2}3 + LiCH3 →UCl{N(SiMe3)2}3 + NaBH4 →

Answer 11.11

UCl{N(SiMe3)2}3 + LiCH3 → U(CH3){N(SiMe3)2}3 + LiCl

UCl{N(SiMe3)2}3 + NaBH4 → U(BH4){N(SiMe3)2}3 + NaCl.

SPH SPH


200

SPH SPH


12 Electronic and MagneticProperties of the Actinides


� understand that the Russell–Saunders coupling scheme is not a good approximation here;� recognize that the electronic spectra of compounds in the +3 and +4 states are dominated

by f–f transitions;� know that transitions are more sensitive to ligand than with the 4f metals;� appreciate that interpretation of spectroscopic and magnetic properties is more difficult

than for the lanthanides.

12.1 Introduction

In general, it is more difficult to interpret the spectra and magnetic behaviour of actinidecompounds than those of lanthanide compounds, so this chapter will provide a qualitativeand somewhat superficial discussion of the phenomena rather than a qualitative one, con-centrating largely upon uranium compounds. The reasons for this is that spin–orbit couplingplays a more important part in actinide chemistry as the 5f orbitals and their electrons arenot so ‘core-like’ as the 4f, particularly in the early part of the actinide series. This meansthat the Russell–Saunders (RS) coupling scheme, which treats spin–orbit coupling as beingmuch weaker than interelectronic repulsion terms, is not applicable in most cases. Neither,however, can one usually apply the other extreme, the j–j coupling scheme, which relieson spin–orbit coupling being strong compared with electrostatic repulsion. Thus the ‘inter-mediate’ coupling scheme (intermediate between RS and jj) is used. It will be recalled thatthe f–f electronic transitions in the spectra of lanthanide complexes are relatively weak incomparison with those of transition metal complexes. However, in the case of the actinides,the 5f orbitals are larger than lanthanide 4f orbitals, so that they interact more with ligandorbitals, causing much higher extinction coefficients and also, because covalency is greater,to create greater nephelauxetic effects in actinide spectra. This means that there is morevariation in both position and intensity of absorption bands than in lanthanide compounds.The ‘forbidden’ electronic dipole transitions are allowed in the presence of an asymmetricligand field, which can arise by either a permanent distortion or by temporary coupling withan asymmetric metal–ligand vibration (vibronic coupling). Apart from the f–f transitions,there are two more types of absorption bands to note in actinide spectra. In general, theparity-allowed 5f–6d transitions occur above 20 000 cm−1, since the 6d levels are con-siderably above 5f for most actinides; these are more intense (and broader) than the f–f


SPH SPH


202 Electronic and Magnetic Properties of the Actinides

transitions. In the case of the free U3+ ion, the 5f 26d1 level is over 30 000 cm−1 above the5f 3 ground state, while in U3+(aq) the charge-transfer transitions start around 24 000 cm−1;solvation thus has a very significant effect upon the relative energies of the 5f and 6d elec-trons. Metal–ligand charge-transfer transitions have their maxima out in the ultraviolet, but,as with transition metals, the tail of these broad and often very intense absorption bandsruns into the visible region of the spectrum, and is responsible for the red, brown, or yellowcolours often noted for actinide complexes with polarizable ligands like Br or I.

12.2 Absorption Spectra

12.2.1 Uranium(VI) – UO22+ – f 0

The ground state of the uranyl ion has a closed-shell electron configuration. There is acharacteristic absorption ∼25 000 cm−1 (400 nm) which frequently gives uranyl compoundsa yellow colour (though other colours like orange and red are not infrequent). This absorptionband often exhibits fine structure due to progressions in symmetric O==U==O vibrations inthe excited state, sometimes very well resolved, sometimes not (Figures 12.1 and 12.2).

It should also be remarked that uranyl complexes tend to emit a bright green fluorescenceunder UV irradiation, from the first excited state. This is used by geologists both to identifyand to assay uranium-bearing minerals in deposits of uranium ores.

40

30

1

2

20

Mol

ar a

bsor

ptiv

ity

10

350 400 450

Wavelength (nm)

500 550

Figure 12.1The absorption spectrum of (1) [UO2(OAc)4]2− in liquid Et4NOAc.H2O, showing the lack of vibronicstructure, due to hydrogen bonding; (2) [UO2(OAc)3]− in MeCN solution, showing the progressiondue to the O==U==O stretching vibration (from J.L. Ryan and W.E. Keder, Adv. Chem. Ser., 1967,71, 335 and reproduced by permission of the American Chemical Society).

SPH SPH


Absorption Spectra 203

400

0.0

0.5

420 440 460 480 500 520 540 560 580 600

Wavelenght (nm)

Abs

orba

nce

2

1

Figure 12.2Absorption spectra of THF solutions of 1 [UO2Cl{η3-CH(Ph2PNSiMe3)2}(thf)] and and 2 [UO2Cl{η3-N(Ph2PNSiMe3)2}(thf)] (reproduced with permission of the Royal Society of Chemistry from M.J.Sarsfield, H. Steele, M. Helliwell, and S.J. Teat, Dalton Trans., 2004, 3443).

12.2.2 Uranium(V) – f 1

The 2F ground state is split into two levels, 2F7/2 and 2F5/2, by spin–orbit coupling in thefree ion. These are split further under the influence of a crystal field; the effect on the energylevels of increasing the crystal field up to the strong field limit is shown in Figure 12.3.

Four transitions are thus expected in the electronic spectrum and generally, in practice,four groups of lines are seen, between the near-IR and the visible, bearing out this prediction.The ground state is a Kramers’ doublet (�7), so UV compounds are EPR active.

Doublet Γ6

Quartet Γ′8

Quartet Γ 8

Doublet Γ′7

Doublet Γ7

2F5/2

2F7/2

t1u(Γ4)

t2u(Γ5)

a2u(Γ2)

Increasing crystal field

Ene

rgy

Figure 12.3The effect of increasing crystal field upon the energy of an electron in an f 1 system such as Uv

(reproduced with permission from S.A. Cotton. Lanthanides and Actinides, Macmillan, 1991, p. 109).

SPH SPH



12.2.3 Uranium(IV) – f 2

The ground state arising from the f2 configuration is 3H4 (Figure 12.4) and the effect ofa crystal field is to split both that and excited states further. A large number of electronictransitions are thus expected, and this is borne out in practice (Figures 12.5 and 12.6).

It will be noted that the transitions are often broader than those found in the spectraof lanthanide complexes – and indeed the later actinides, see Section 12.2.4. The 5f en-ergy levels are more sensitive to coordination number than are the corresponding levelsin the lanthanides; since there are bigger crystal-field effects, one sees pronounced differ-ences between the spectra of 6-coordinate [UCl6]2− and of U4+(aq) (Figure 12.5), leadingto the conclusion that the uranium(IV) aqua ion was not six coordinate (most recent EX-AFS results suggest a value of 9 or 10, see Table 11.1). Figure 12.6 displays anotherexample of the difference in spectra between similar complexes of different coordinationnumber.

1G4

1G

1D2

1D

T 2a

T 1g

Eg

Eg

EgE

Crystal field

A 1g

D 4hO h

A 2g

A 2g

A 2gB 1g

B 2g

3 F4

3 P1

3 P

1S

1I

3 P2

3 P0

1I6

1 S0

3 F

3 H

3 F33 F2

3 H6

3 H5

3 H4

∆

Spin−orbit coupling

Electrostatic repulsion

f 2Figure 12.4A qualitativeenergy-leveldiagram for the U4+

ion, showingsuccessively theeffects ofelectrostaticrepulsion,spin–orbitcoupling, andcrystal-fieldsplitting (the lattershown only for theground state).Overlap betweenlevels is neglected.Adapted from M.Hirose et al., Inorg.Chim. Acta, 1988,150, L93, andreproduced bypermission of theEditor.

SPH SPH


Absorption Spectra 205

400 500 600 700 800

Wavelength (nm)

Abs

orpt

ion

(arb

itrar

y)

900 1000 1100

A

B

Figure 12.5The absorption spectra of octahedral [UCl6]2− (A) and 9–10-coordinate U4+(aq) (B) (redrawn fromD.M. Gruen and R.L. Macbeth, J. Inorg. Nucl. Chem., 1959, 9, 297 and reproduced by permission ofElsevier Science Publishers).

Figure 12.6Solid-state absorption spectra of octahedral [UCl4(But

−2SO)2] (A) and 8-coordinate [U(Me2SO)2] I4

(B) (redrawn from J.G.H. DuPreez and B. Zeelie, Inorg. Chim. Acta, 1989, 161, 187 and reproducedby permission of Elsevier Science Publishers).

SPH SPH



Detailed investigations have been made of the octahedral [UCl6]2− ion. Its spectrumis largely vibronic in nature, with electronic transitions accompanied by vibrations of thecomplex ion (odd-parity modes – the T1u asymmetric stretch and T1u and T2u deformations).Here, as in other UIV cases, overlap of bands from different states occurs because of thesimilarity in crystal-field and spin–orbit coupling effects. Its spectrum can be altered by de-stroying the centre of symmetry (e.g., by hydrogen bonding), which enables pure electronictransitions to be observed, and alters band patterns in multiplets.

12.2.4 Spectra of the Later Actinides

Because of the relatively short half-lives of many later actinides, purity of samples andcorrect identification of lines can be a matter of uncertainty, but Figure 12.7 shows how this

Figure 12.7The absorption spectrum of the hexagonal form of BkCl3 as a function of time. The changes in thespectrum are due to the formation of CfCl3. Note the sharp ‘lanthanide-like’ transitions characteristicof the later actinide (+3) state. (from J.R. Peterson et al., Inorg. Chem., 1986, 25, 3779 reproducedby permission of the American Chemical Society).

SPH SPH


Magnetic Properties 207

can be turned to advantage. It shows spectra obtained over a period of time from a sampleof 249BkCl3 beginning 11 days after synthesis. Now, 249Bk is a β-emitter with a half-lifeof 320 days, and the spectrum obtained over a 976 day period (three half-lives, duringwhich time the berkelium decays to some 12.5 % of its original amount) shows the lossof the characteristic absorptions due to the 249BkCl3 and their replacement by a spectrumdue to 249CfCl3. These spectra are very reminiscent of the sharp, line-like absorptionsobtained from the lanthanides. This reflects the fact that chemically the heavy actinides arelanthanide-like, suggesting that with increasing atomic number the 5f orbitals are now morecore-like and thus less readily influenced by environment.

After the emission of a β-particle from the Bk nucleus the californium ion regains anelectron to maintain the (+ 3) oxidation state:

24997Bk3+ → 249

98Cf 4+ + e−

Cf 4+ + e− → Cf 3+

It may also be noted that crystal type is retained; X-ray diffraction confirms that the CfCl3retains the hexagonal structure of the original BkCl3 rather than adopting the orthorhombicmodification.

12.3 Magnetic Properties

Uranium(VI) compounds are expected to be diamagnetic, with their 1S0 (f 0) groundstate. However, compounds like UF6 and uranyl complexes in fact exhibit temperature-independent paramagnetism, explained by a coupling of paramagnetic excited states withthe ground state.

Uranium(V) compounds are, as expected for an f 1 system, paramagnetic, usually ex-hibiting Curie–Weiss behaviour, with large Weiss constants; g-values, expected to be 6/7,are modified by the mixing in of higher states and by orbital-reduction effects (covalency),experimental g-values including values of 1.2 in Na3UF8 and 0.71 in CsUF6.

Matters are more complicated for uranium(IV); this f 2 system has a 3H4 ground state,the energy level diagram has already been given (Figure 12.4). In a regular octahedralgeometry, there is no contribution to the paramagnetic susceptibility from the first-orderZeeman term. Species like the [UCl6]2− ion (and isoelectronic PuF6) display temperature-independent paramagnetism, caused by the second-order Zeeman term mixing the 3T1g

excited state into the ground state. In lower symmetry, such as a D4h trans-UX4L2 complex,both the first- and second-order Zeeman effects contribute to the susceptibility. If thereis a small distortion from a regular octahedron, the splitting of the 3T1g excited state issmall, so that �E in Figure 12.4 is large. There is thus little thermal population of theseexcited states, so the first-order Zeeman effect is small; the paramagnetic susceptibilityshows little or no temperature dependence. As the distortion becomes larger, in complexeslike UBr4(Et3AsO)2and UI4[(Me2N)3PO)2], thermal population of a component of the 3T1g

excited state becomes more feasible, and thus the susceptibility shows a greater temperaturedependence. In the case of a cis-UX4L2 complex, with D4h symmetry, there is no first-order Zeeman term, so that the second-order Zeeman effect causes temperature-independentparamagnetism.

Figure 12.8 shows variable-temperature susceptibility data for some UIV complexes ofthese types, which is in keeping with these explanations.

SPH SPH



50

D

C

B

2

3

4

5

6

7

8

9

A

100 150 200 250 300 350Temperature (K)

χ(m

ol ×

103 /

cgs

emu)

Figure 12.8Temperature dependence of the magnetic susceptibility of some uranium(IV) complexes: (A) trans-[UBr4(Et3AsO)2]; (B) trans-[UCl4(Et3AsO)2]; (C) (Ph4P)2 [UCl6]; (D) cis-[UCl4(Ph3PO)2] (redrawnfrom B.C. Lane and L.M. Venanzi, Inorg. Chim. Acta, 1969, 3, 239 and reproduced by permission ofthe editor).

Few data are available for uranium(III) compounds, but a number of compounds withthe f 3 configuration, like Cs2NaUCl6, have µ ∼ 3.2 µB, which is largely temperatureindependent.

SPH SPH


13 Organometallic Chemistryof the Actinides


� recall that these compounds are very air- and water-sensitive;� recall that the bonding has a significant polar character, especially in the +3 state, but

that +4 compounds, especially those of uranium, have more stability;� recall that these compounds usually have small molecular structures with appropriate

volatility and solubility properties;� recall appropriate bonding modes for the organic groups;� suggest suitable synthetic routes;� suggest structures for suitable examples;� appreciate the contribution of the C5Me5 ligand in uranium chemistry;� recall the unique properties of uranocene;� appreciate the unusual compounds that can be made using imide and CO as ligands.

13.1 Introduction

Although lacking π -bonded compounds in low oxidation states that characterize the d-blockelements, the actinides have a rich organometallic chemistry. Their compounds frequentlyexhibit considerable thermal stability, but like the lanthanide compounds are usually in-tensely air- and moisture-sensitive. They are often soluble in aromatic hydrocarbons suchas toluene and in ethers (e.g., THF) but are generally destroyed by water. Sometimes theyare pyrophoric on exposure to air. Most of the synthetic work has been carried out with Thand U; this is partly due to the ready availability of MCl4 (M = Th, U) and also because ofthe precautions that have to be taken in handling compounds of other metals, especially Puand Np.

13.2 Simple σ-Bonded Organometallics

During the Manhattan project, a wide range of compounds was screened in the search forvolatile uranium compounds that could be used for separation of isotopes of uranium, sinceUF6 was not an easy compound to handle. The leading organometallic chemist of the day,Henry Gilman, investigated the synthesis of simple alkyls and aryls, concluding that suchcompounds did not exist, or at least were unstable. Even today, just one neutral homoleptic


SPH SPH


210 Organometallic Chemistry of the Actinides

alkyl of known structure exists, [U{CH(SiMe3)2}3], made from an aryloxide by a routeensuring no contamination with lithium halide:

3 LiCH(SiMe3)2 + U(O-2, 6-But2C6H3)3 → [U{CH(SiMe3)2}3]

+ 3 Li(O-2, 6-But2C6H3)

If LiCH(SiMe3)2 is treated directly with UCl3(THF)3 in a salt-elimination reaction thealkylate salt [(THF)3Li-Cl-U{CH(SiMe3)2}3] is obtained.

[U{CH(SiMe3)2}3] has a trigonal pyramidal structure, like the corresponding lanthanidealkyls (Section 6.2.1) and also like the isoelectronic amides [M{N(SiMe3)2}3] (M = U,lanthanides). There are some agostic U . . . . H–C interactions. Similar Np and Pu compounds[M{CH(SiMe3)2}3] have been made. Ab initio MO calculations for hypothetical M(CH3)3

(M = U, Np, Pu) molecules also predict pyramidal structures as a result of better 6dinvolvement (not 5f ) in the M–C bonding orbitals in pyramidal geometries.

The benzyl [Th(CH2Ph)4] has been synthesized and the pale yellow dimethylbenzyl com-pound Th(1,3,5-CH2C6H3Me2)4 has also been reported, but their structures are not known(and may have multihapto-coordination of benzyl to thorium). The heptamethylthorate an-ion [ThMe7]3− is found in the salt [Li(tmeda)]3[ThMe7] (tmeda = Me2NCH2CH2NMe2).The latter contains thorium in monocapped trigonal prismatic coordination; one methyl isterminal (Th–C 2.571 A) whilst the other six methyl groups pairwise bridge the lithium andthorium (Th–C 2.655–2.765 A). The methyl groups are, however, equivalent in solution(NMR).

Certain simple alkyls are stabilized by the presence of either phosphine or amide ligands,thus:

UCl{N(SiMe3)2}3 + CH3Li → U(CH3){N(SiMe3)2}3 + LiCl

2 ThCl{N(SiMe3)2}3 + Mg(CH3)2 → 2 Th(CH3){N(SiMe3)2}3 + MgCl2

These are pentane-soluble, volatile tetrahedral molecules. They have an extensive chemistryinvolving metallacycles which will be dealt with later (Section 13.6). Although MMe4

(M = Th, U) have not been isolated, this unit can be stabilized as eight-coordinate phosphineadducts, where the role of phosphines is to saturate the coordination sphere. The coordinationof a ‘soft’ phosphine ligand to a ‘hard’ metal ion like uranium(IV) is noteworthy.

MCl4(dmpe)2 + 4 CH3Li → M(CH3)4(dmpe)2 + 4 LiCl

These react with phenol forming phenoxides:

Th(CH3)4(dmpe)2 + 4 C6H5OH → Th(OC6H5)4(dmpe)2 + 4 CH4

Similar compounds [M(CH2Ph)4(dmpe)2] (stable to 85 ◦C in the absence of air) and[M(CH2Ph)3Me(dmpe)2] have been isolated; structures of [Th(CH2Ph)4(dmpe)2] and[U(CH2Ph)3Me(dmpe)2] confirm their identity, they have short M–C contacts. Compoundsof alkyl groups like CH2CH3 that are capable of undergoing elimination have not beenisolated.

Compounds containing one or more cyclopentadienyl-type ligands are much morerobust. (Throughout this section, C5H5 is abbreviated to Cp; C5Me5 is abbreviated toCp*; C5H4.CH3 is MeCp.)

SPH SPH



13.3 Cyclopentadienyls

These are regarded as derivatives of the C5H5− ligand; substituted rings also occur. Some

of these were the first organoactinide compounds to be reported, shortly after the discoveryof ferrocene. The most important compounds are in the +4 state but there is a significantchemistry of U(+3) as well as examples in oxidation states (+5) and (+6). One factor thatshould be noted is that replacing one or more hydrogens in the cyclopentadienyl ring can havea significant effect upon the stability and isolability of complexes made from them, throughboth steric and electronic effects. Thus three or four Cp ligands can be accommodatedaround uranium(IV), whereas with Cp* two is the norm. In another example, [UCp3] givesno sign of reacting with carbon monoxide; [U{C5H3(SiMe3)2}3] forms a compound thatcan be detected in solution; and [U(C5Me4H)3] forms [U(C5Me4H)3(CO)], which can beisolated in the solid state.

13.3.1 Oxidation State (VI)

High oxidation states are not normally associated with organometallic chemistry, but aremarkable uranium(VI) imide has been made (Figure 13.1):

UCp∗2(CH3)Cl + Li[PhNN(H)Ph] → UCp∗

2(= NPh)2 + CH4 + LiCl.

The reaction may go via an intermediate [UCp*2(η2-PhN–NPh)]. There are no f–f transi-tions in the visible spectrum, confirming its UVI-f 0 character, whilst the structure of thisremarkable compound shows the expected pseudotetrahedral geometry; the short U–N bondlength (1.952 A) confirms its multiple-bond character and the U==N–Ph linkage is virtuallylinear.

NPh

NPh

U

Figure 13.1A uranium(VI) imide.

13.3.2 Oxidation State (V)

Rare organoimides U(MeCp)3(=NR) have been made by oxidation of U(MeCp)3(thf) withRN3 (R = Me3Si and Ph). The U–N bond length of 2.109 A is regarded as having multiple-bond character (cf. the U=O bond in uranyl compounds).

U(MeCp)3(THF) + RN3 → U(MeCp)3(==NR) + THF + N2

13.3.3 Oxidation State (IV)

There are three families of compounds whose structures can be regarded as derivedfrom a tetrahedral UX4 molecule with one or more halogens replaced by one or more

SPH SPH



cyclopentadienyl rings; [Cp4An], [Cp3AnX]; [CpAnX3] (X = halogen, most usuallyCl). Since halogens have less steric influence than a cyclopentadienyl group, a CpAnX3

molecule would be coordinatively unsaturated, so CpAnX3 actually exist as solvates likeCpAnX3(THF)2. Cp2AnX2are unknown for X = halogen, attempts to prepare them resultingin mixtures of [Cp3AnX] and [CpAnX3], but they are known for X = NEt2 and BH4.

Syntheses

UCl4 + 4 KCp → [UCp4] + 4 KCl (C6H6)

UCl4 + 3 TlCp → [UCp3Cl] + 3 TlCl (1,2-dimethoxyethane)

UCl4 + TlCp → [UCpCl3(THF)2] + TlCl (THF)

Some similar compounds can be made with other actinides, namely [MCp4] (M = Th, Pa,Np), [MCp3Cl] (M = Th, Np), and [ThCpCl3].

Sometimes other starting materials are used:

U(NEt2)4 + 3 CpH → [UCp3(NEt2)] + 3 HNEt2

O

Cl

ClU

O

Cl

MClM

Figure 13.2Structures of MCp4, MCp3Cl and MCpCl3 (THF)2.

Structures

[MCp4] have regular tetrahedral structures whilst [MCp3Cl] are analogous (Figure 13.2)with the halogen occupying a position corresponding to the centroid of a fourth ring and[UCpCl3(thf)2] has a pseudo-octahedral structure.

[MCp3Cl] are the most useful of these compounds as the halogen can be replaced by anumber of groups (e.g., NCS, BH4, acac, Me, Ph) (Fig. 13.3).

UCl4NaCp

Cp3UCl

KNO3(aq)

[Cp3U(OH2)2]+ NO3−

MeLi NaOMe

NaBH4

MSCN (M = K, Na)

LiPPh2

NaRuCp(CO)2

Cp3UMe Cp3U(OMe)

Cp3U(BH4)

Cp3U(SCN)

Cp3U(PPh2)

Cp3URuCp(CO)2

Figure 13.3Synthesis and reactions of Cp3UCl.

SPH SPH



There is evidence for a significant covalent contribution to the bonding in some of theseactinide(IV) compounds; UCp3Cl does not react with FeCl2 forming ferrocene, whereasMCp3 do (M, e.g., Ln, U). It appears to undergo ionization in aqueous solution as a stablegreen [UCp3(OH2)n]+ (n ∼ 2) cation. There is also some evidence from Mossbauer spectraof neptunium-(III) and -(IV) cyclopentadienyl compounds that in a NpIV compound such asNpCp4 (but not NpIII compounds) there is a greater shielding of the 6s shell than in an ioniccompound like NpCl4, leading to enhanced covalence in the bond. Bonds are reasonablystrong; the mean bond dissociation energy in UCp4 has been determined to be 247 kJ mol−1,compared with a value of 297 kJ mol−1 for FeCp2.

The most interesting [MCp3X] compounds are the alkyls and aryls, [MCp3R] being madeby reaction with either organolithium compounds or Grignard reagents.

[MCp3Cl] + RLi → [MCp3R] + LiCl; [MCp3Cl] + RMgX → [MCp3R] + MgXCl

(M = Th, U; R, e.g., CH3, Prn, Pri, Bun, But, C6H5, C6F5, C≡CPh, CH2Ph, etc.)

They were independently synthesized by three independent research groups at the start ofthe 1970s and represent the first compounds to have well-defined actinide-carbon σ bonds(structures for M = U; R, e.g., Bu, CH2C6H4CH3, etc). (See, e.g., the papers by T.J. Marksgroup in J. Am. Chem. Soc., 1973, 95, 5529; 1976, 98, 703.) Thermally stable in vacuo,they react with methanol, forming MCp3(OMe) and RH. A low-temperature NMR study of[UCp3Pri] indicates restricted rotation about the U–C σ bond at low temperatures.

Bond disruption energies of 315–375 kJ/mol for the Th–C bonds in [ThCp3R] (R, e.g.,CH3, CH2Ph) indicate the thermodynamic stability of these compounds, comparable withtransition metal alkyls (the bond energies are 10–30 kJ/mol less for the uranium compounds).Decomposition does not occur by theβ-elimination route; instead intermolecular abstractionof a hydrogen from a Cp ligand occurs (Figure 13.4) forming RH. A stable compound[Cp2Th(µ-C5H4)2ThCp2] is the other product (Figure 13.5).

Th

R

HTh

R

Figure 13.4Decomposition of ThCp3R.

These compounds also undergo insertion reactions of CO, CO2, and R′NC into the M–Csigma bond, forming [Cp3M(η2-COR)], [Cp3M(η2-O2CR)], and [Cp3M(η2-C(R)NR′], re-spectively (Figure 13.6).

ThTh

Figure 13.5Structure of [Cp2Th(µ-C5H4)2ThCp2].

SPH SPH



Cp3MR

CO

CO2

Cp3M

O

R

Cp3M

O

O

R

R'NC

Cp3M

N

R

R'

Figure 13.6Insertion reactions of Cp3MR.

As already mentioned, there is no UCp2Cl2 – attempts to make it giving mixtures of[UCp3Cl] and [UCpCl3], but some UCp2X2 can be made for example:

[U(NEt2)4] + 2 C5H6 → UCp2(NEt2)2 + 2 HNEt2UCl4 + 2 NaBH4 → UCl2(BH4)2 + 2 NaCl followed by

UCl2(BH4)2 + 2 TlCp → UCp2(BH4)2 + 2 TlCl

UCp2(NEt2)2 reacts with ROH [R = bulky group, e.g., [(But)3C; 2,6-Me2C6H3]:

UCl2(NEt2)2 + 2 ROH → UCp2(OR)2 + 2 HNEt2

13.3.4 Oxidation State (III)

A variety of syntheses are possible to these compounds, including the expected salt-elimination route:

UCl3 + 3 NaCp → UCp3 + 3 NaCl (C6H6)

Reduction with sodium naphthalenide is often possible:

MCp3Cl + Na → MCp3 + NaCl (in THF) (M = U, Th, Np)

An unusual route used for several of the transuranium elements is a microscale reaction ofthe metal trichlorides with molten BeCp2 at 65 ◦C.

2 MCl3 + 3 BeCp2 → 2 MCp3 + 3 BeCl2 (M = Pu, Am, Cm, Bk, Cf)

The cyclopentadienyl products were sublimed out of the reaction mixture and characterizedby powder diffraction data, showing they were isostructural with the corresponding LnCp3

(Ln = Pr, Sm, Gd).Similar compounds have been made with substituted cyclopentadienyl groups; their prop-

erties strongly resemble those of the corresponding LnCp3. The unsolvated U(C5H4SiMe3)3

is known to have trigonal planar geometry, as is U(C5Me4H)3 (X-ray). Lewis bases, e.g.THF, C6H11NC, py, form pseudotetrahedral adducts UCp3.L {X-ray, e.g. for [UCp3(THF)]}in the same way that lanthanides do. Some adducts are known with rather soft bases liketertiary phosphines, e.g [U(MeCp)3(PMe3)], that might not have been expected for a fairly‘hard’ metal such as UIII.

SPH SPH


Compounds of the Pentamethylcyclopentadienyl Ligand (C5Me5 = Cp*) 215

Evidence from UV–visible spectra (Cm and Am compounds) and Mossbauer spectra(Np compounds) indicates that these compounds have probably rather more covalent con-tributions to their bonding than do the corresponding lanthanide compounds, but are largelyionic compared with cyclopentadienyls in the +4 state.

Analogous compounds AnCp3 exist for other actinides, e.g. Th, Np (?), Pu, Am, Bk, Cf,Cm. The thorium(III) compound is especially noteworthy, on account of the rarity of thisoxidation state; the structure of the similar Th{C5H3(SiMe3)2}3 is known to have trigonalplanar geometry around thorium.

3 Th{C5H3(SiMe3)2}2Cl2 + Na/K(alloy) → Th + 2 Th{C5H3(SiMe3)2}3 + 6 Na/KCl

These ThIII compounds are EPR-active; spectra indicate a 6d1 ground state forTh{C5H3(SiMe3)2}3, despite the fact that the free Th3+ ion has a 5f 1 configuration.

13.4 Compounds of the Pentamethylcyclopentadienyl Ligand (C5Me5 = Cp*)

13.4.1 Oxidation State(IV)

Compounds of the type [MCp*2X2] (M = Th, U) are rather important. They owe theirimportance to the fact that although three Cp* rings can be accommodated round uraniumin UCp*3Cl,this compound cannot be made by direct synthesis (nor can MCp*3), partlyowing to the bulk of the C5Me5 ligands, and the [MCp*2X2] systems are in practice morereadily obtained.

MCl4 + 2 Cp∗MgBr → MCp∗2Cl2 + 2 MgBrCl (M = Th, U) in toluene

Halides can be replaced by suitable ligands, thus reactions with NaOMe and LiNMe2

afford products like ThCp*2(OMe)2, ThCp*2Cl(NMe2), and ThCp*2(NMe2)2. Li2S5 reactsforming [ThCp*2(S5)]. Importantly, the chloride compounds can be alkylated and arylated.Either one or two chlorines can be replaced.

MCp∗2Cl2 + LiR → MCp∗

2(R)Cl + LiCl

MCp∗2(R)Cl + LiR → MCp∗

2R2 + LiCl

(M = Th, U; R, e.g., Me, Ph, CH2CMe3, CH2SiMe3)

They have the usual pseudotetrahedral coordination; the structure of Cp*2Th(CH2CMe3)2

reveals agostic Th....H–C interactions with the α-C–H bonds. The M–C σ bonds seemto have a significant ionic contribution; thus ThCp*2Me2 reacts with acetone to afford atert-butoxy compound.

Reactions with R′OH, R′SH, and R22NH afford products such as ThCp*2(R)(OR′),

ThCp*2(OR′)2, ThCp*2(SR′)2, and ThCp*2(NR22)2(R′, e.g., Pr; R2, e.g., Et).

The compounds Cp*2Th(CH2CMe3)2 and Cp*2Th(CH2SiMe3)2 undergo most remark-able elimination reactions to form metallacyclic complexes on heating to only 50 ◦C intoluene solution:

|−−−−−−−−−−−|Cp∗

2Th(CH2SiMe3)2 → Cp∗2Th(CH2CMe2-CH2) + CMe4

|−−−−−−−−−−−−|Cp∗

2Th(CH2SiMe3)2 → Cp∗2Th(CH2SiMe2−CH2) + CMe4

|−−−−−−−−−−−|Reactions of Cp*2Th(CH2CMe2-CH2) are shown in Figure 13.7.

SPH SPH



+ CMe4Th

CH2 But

CH2ButCH2But

CH2 But

50°C

heptane

Th

ThCH3

ThH

H

H2

CH4ROH

ThORCp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Th

Figure 13.7 |−−−−−−−−−−−−−|Reactions of Cp*2Th(CH2CMe2-CH2).

Some mono(Cp*) complexes are known, reaction of MCl4 (M = U, Th) withMeMgCl.THF yielding [Cp*MCl3(thf)2], with structures analogous to the [CpLnCl2(thf)3]compounds. The halogens can be replaced with the isolation of [Cp*Th(CH2Ph)3]; this hasa piano-stool structure with Th–C σ -bonds of 2.578–2.581 A. There is some interactionbetween Th and the benzene rings, so this is not a classical η1-alkyl.

13.4.2 Cationic Species and Catalysts

Cationic alkyls have been studied with profit. Thus, in benzene solution,

Cp∗2Th(CH3)2 + [HNBun

3]3+[B(C6F5)4]− → [Cp∗

2Th(CH3)]+[B(C6F5)4]−

+ NBun3 + CH4

[Cp*2Th(CH3)]+ [B(C6F5)4]− in fact has a very weakly bound anion [Th–F 2.757, 2.675A; Th–C (ring) 2.754 A, Th–CH3 2.399 A], so that it is an active catalyst for ethenepolymerization and hex-1-ene hydrogenation (Figure 13.8).

If this type of abstraction reaction is carried out in THF, [Cp*2Th(CH3)(THF)2]+ [BPh4]−

is formed [Th–C (ring) 2.801 A , Th–CH3 2.491 A].

+H2

−CH4

−

ThCp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*

Cp*CH3 Th H

Th

Th

HH

+

+ H2

+ +

+

+

Figure 13.8[Cp*2ThCH3]+ as a catalyst for hydrogenation.

SPH SPH


Compounds of the Pentamethylcyclopentadienyl Ligand (C5Me5 = Cp*) 217

CH3CH3

CH3

CH3

CH3

Th

Thdehydroxylatedalumina

partlydehydroxylatedalumina

ThO

Al

Figure 13.9Surface bound cationic thorium alkyls.

On alumina surfaces cationic alkyls exhibit catalytic activity, due to the formation ofcoordinatively unsaturated species (Figure 13.9).

13.4.3 Hydrides

Striking hydride derivatives are formed, which are efficient catalysts for the homogeneoushydrogenation of hex-1-ene (Figure 13.10):

2 ThCp∗2R2 + 4 H2 → Cp∗

2Th(H)(µ-H)2Th(H)Cp∗2 + 4 RH R, e.g., Me

This remarkable thorium compound is stable to 80 ◦C. In solution it has rapid hydrogenexchange with dihydrogen gas. Solution 1H NMR shows the bridge and terminal hydrogensare equivalent even at −90 ◦C. In the solid state, Th–H is 2.03 A (terminal) and 2.29 A(bridge); a Th–Th distance of 4.007 A indicates minimal metal–metal bonding. In the IR,ν Th–H (terminal) vibrations occur at 1404 and 1336 cm−1; bridging vibrations occur at1215, 1114, 844, and 650 cm−1. Cp*2U(H)(µ-H)2U(H)Cp*2 can be made too but tends to

Th

H

H

H

ThH

Figure 13.10Structure of [Cp*2Th(H)(µ-H)2Th(H)Cp∗

2].

SPH SPH



decompose to a UIII hydride Cp*2U(H), which can be isolated and stabilized as a DMPEcomplex [Cp*2U(H){Me2P(CH2)2PMe2}].

Attempted conversion of UCp*2RCl into a hydride gives the uranium (III) compoundUCp*2Cl (Section 13.4.4).

The hydrides are reactive and readily react with halogenocarbons, alcohols, and ketones;alkenes afford alkyls.

13.4.4 Oxidation State (III)

Hydrogen reduction of [UCp*2ClR] gives trimeric UCp*2Cl (Figure 13.11).

UCp∗2Cl(CH2SiMe3) + H2 → UCp∗

2Cl + SiMe4

U

Cl Cl

ClU U

Figure 13.11The structure of trimeric [Cp*2UCl]3.

In the similar reaction with R = methyl, the organic product is CH4. If the reaction iscarried out with the analogous thorium starting material, ThCp*2Cl(R) forms the thorium(IV)compound [Cp*2Th(µ-H)Cl]2.

The uranium compound has a cyclic trimeric structure. Each uranium is individually inpseudotetrahedral coordination. It undergoes various reactions (Figure 13.12).

These may be classed as straightforward substitutions using a bulky anionic ligand;adduct formation with Lewis bases retaining pseudotetrahedral coordination; and redox.The photoelectron spectrum of the alkyl [UCp*2(CH2SiMe3)] indicates a 5f 3 ground statebut also that the 5f 26d1 state is only slightly higher in energy.

[Cp*2UCl]3

MR

L[Cp*2UCl(L)]

[Cp*2UCl2]

[Cp*2U(CH3)Cl]

L = py, Et2O, THF, PMe3

[Cp*2UR] R = CH(SiMe3)2 ; N(SiMe3)2

CH3Cl [Cp*2UCl2] andC6H5Cl

Figure 13.12Reactions of [Cp*2UCl]3.

SPH SPH


Cyclooctatetraene Dianion Compounds 219

13.5 Tris(pentamethylcyclopentadienyl) Systems

For many years it was believed that [LnCp*3] systems could not be made. After the success-ful isolation of [SmCp*3] and similar compounds (Section 6.3.2), attention turned to thepossibility of isolating the uranium analogue. The hydride Cp*2UH(dmpe) proved to be auseful starting point, reacting with either tetramethylfulvalene or PbCp*2 to form [UCp*3],but routes analogous to those used for lanthanides have also worked (Figure 13.13).

KBPh4BU + KCp* U +

+UH

P

P

U + DMPE

Figure 13.13Synthesis of [UCp*3].

With the isolation of [UCp*3] accomplished, the next target was what promised to beeven harder, the synthesis of the uranium(IV) species [UCp*3Cl] and related compounds.[UCp*3] underwent oxidation with HgF2, forming [UCp*3F], and reaction with a stoichio-metric amount of PhCl, forming [UCp*3Cl]; reaction of [UCp*3] with 2 molecules of PhClafforded [UCp*2Cl2].

The structure of [UCp*3] has trigonal planar coordination of U; similarly in [UCp*3Cl];the three ring centroids are coplanar with U, and with a very long U–Cl bondat 2.90 A – compare 2.637 A in [U(C5Me4H)3Cl].

13.6 Other Metallacycles

|−−−−−−−−−−−−−−−|Cp*2Th(CH2CMe2-CH2) has already been mentioned. Another kind of metallacycle hasbeen synthesized from the alkyls [M(CH3){N(SiMe3)2}3] (M = Th, U, see Section 11.8.1).They undergo thermolysis in benzene solution to afford [{(Me3Si)2N}2M{N(SiMe3)(SiMe2)}]. These in turn have a rich chemistry (Figure 13.14), undergoing insertion re-actions with CO, carbonyl compounds, nitriles, and isocyanides. In the reactions involvingcompounds with acidic hydrocarbons, such as tertiary alcohols, phenols, and thiols, themetallacyclic ring is broken to re-form the amide ligand.

13.7 Cyclooctatetraene Dianion Compounds

These are regarded as derivatives of the C8H82− ligand; substituted rings also occur. Ura-

nocene, U(η8-C8H8)2, was reported in 1968 (A. Streitweiser, Jr., and V. Mueller-Westerhoff,

SPH SPH



{(Me3Si)2N}2M

{(Me3Si)2N}2M

N

SiMe3

SiMe3

SiMe3

SiMe3

SiMe2

SiMe2

SiMe2

SiMe2

[M(SR){N(SiMe3)2}3]

[M(Me){N(SiMe3)2}3]

heat

ArSH

ROH

[M(OR){N(SiMe3)2}3]

[M(NEt2){N(SiMe3)2}3]

Et2NH

{(Me3Si)2N}2M

{(Me3Si)2N}2M

N

O

N

O

R R'

CO

RR'CO

N

N

ButButNC

Figure 13.14Synthesis and reactions of the metallacycles [{(Me3Si)2N}2M{N(SiMe3)(SiMe2)}].

J. Am. Chem. Soc., 1968, 90, 7364) and was a seminal compound in the development of theorganometallic chemistry of the actinides. It was made by a general route:

MCl4 + 2 K2C8H8 → M(η8-C8H8)2 + 4 KCl (M = Th, Pa, U, Np)

The Pu analogue was made from (pyH)2PuCl6, since plutonium does not form a tetrachlo-ride. All five compounds are isomorphous and believed to have similar sandwich structures,with the two (planar) 8-membered rings in the eclipsed configuration (Figure 13.15).

U

Figure 13.15Structure of uranocene, [U(C8H8)2].

Uranocene forms green crystals that are pyrophoric in air, but it is otherwise ratherunreactive, not readily hydrolysed by water or even acetic acid. It is stable to vacuumsublimation at 200 ◦C. The other M(η8-C8H8)2 are more reactive and easily hydrolysed.A mean bond-dissociation energy of 347 kJ mol−1 has been measured for U(η8-C8H8)2;it is substantially greater than the value of 247 kJ mol−1for U(C5H5)4. Of substituteduranocenes, two stand out. [U(1,3,5,7-Me4C8H4)2] exists in the crystal as a mixture of the‘eclipsed’ and ‘staggered’ rotamers, whilst [U(1,3,5,7-Ph4C8H4)2] (present in the crystalwith phenyls staggered in the two different rings) sublimes at 400 ◦C, 10−5 mmHg, and istotally air stable, doubtless owing to the phenyl groups hindering access of oxygen moleculesto the uranium atom.

Some sandwich compounds [K(solvent)] [M(η8-RC8H7)2 (M = U, Np, Pu, Am; R, e.g.,H, Me) may also be made in the +3 state by analogous reactions starting from:

MCl3 + 2 K2(RC8H7)Diglyme−−−−→[K(diglyme)][M(η8-RC8H7)2] + 3 KCl

SPH SPH


Arene Complexes 221

+

++

++

+

+

−−

−

−

−

−

−−

Figure 13.16Possible f-orbital involvement in the bonding in uranocene.

Bonding in uranocene has been the subject of controversy for some 30 years. It wasearly on pointed out that orbital symmetry interactions, involving f orbitals, could be drawnanalogous to those in ferrocene (Figure 13.16).

More sophisticated MO calculations, especially including relativistic corrections, suggestthat whilst uranium 6d orbitals interact with the ligand orbitals, the 5f orbitals are relativelyunperturbed. Photoelectron spectra do suggest that 5f interaction increases as more alkylgroups are introduced into the cyclooctatetraene rings.

Half-sandwich compounds have been made, in the case of thorium by redistribution:

Th(η8-C8H8)2 + ThCl4 → 2 (η8-C8H8)ThCl2(THF)2 (in THF)

This compound has a piano-stool-type structure (Figure 13.17).

O

Th

Cl

Cl

O

Figure 13.17Structure of the ‘half-sandwich’ [Th(C8H8)Cl2(THf)2].

Uranium analogues can be made, not always by identical routes:

C8H8 + 2 NaH + UCl4 → (η8-C8H8)UCl2(THF)2 + 2 NaCl + H2 (in THF)

U(η8-C8H8)2 + I2 → (η8-C8H8)UI2(THF)2 + C8H8 (in THF)

13.8 Arene Complexes

The reaction between UCl4, Al, and AlCl3 in refluxing benzene affords [(η6-C6H6)U{(µ-Cl)2AlCl4}3]. (Figure 13.18).

Certain other compounds of this type are known in the (+3) state, such as [(η6-C6H3Me3)U(BH4)3]; also dimeric [(C6Me6)2U2Cl7]+ (AlCl4)−, which is [(η6-C6Me6)Cl2U(µ-Cl)3UCl2(η6-C6Me6)]+ (AlCl4)−. These are considerably more reactive than ura-nocene.

SPH SPH



Al ClClCl

ClCl Cl

Al

Cl Cl

UClClAlCl

Cl

Figure 13.18

However, in contrast to the lanthanides, no arene complexes in a formal zero oxidationstate have been isolated.

13.9 Carbonyls

The early transition metals are conspicuous in forming binary carbonyls like M(CO)6 (M =Cr, Mo, W); M′(CO)5 (M′ = Fe, Ru, Os); and Ni(CO)4, which have considerable thermalstability.

Co-condensation of uranium atoms with CO in an argon matrix at 4 K followed byannealing gives carbonyls U(CO)n detectable by IR, but which decompose above 20 K (asdo those of the lanthanides). A complex identified as U(CO)6 has νCO at 1961 cm−1, verysimilar to that in W(CO)6 (1987 cm−1), a significant decrease from the value of 2145 cm−1

in CO itself, indicating a substantial amount of π -donation by uranium.However, unlike the lanthanides, some stable carbonyl complexes in normal oxidation

states have been identified for uranium. No adduct has been identified for [UCp3] butU{C5H4(SiMe3)}3 reversibly combines with CO in both the solid state and hexane solution,the latter with a colour change from deep green to burgundy, and the appearance of νCO

at 1976 cm−1 in the IR spectrum. [U{C5H3(SiMe3)2}3] forms a similarly detectable butnon-isolable CO adduct; however, an adduct [U(C5Me4H)3(CO)] has been isolated in thesolid state, starting from [U(C5Me4H)3]:

[U(C5Me4H)3] + CO � [U(C5Me4H)3(CO)]

It has a linear U–CO linkage, with U–C 2.383 A, and C–O slightly lengthened from 1.128 Ain free CO to 1.142 A. νCO occurs at 1880 cm−1 in [U(C5Me4H)3(CO)], shifted to 1840 cm−1

in [U(C5Me4H)3(13CO)] and 1793 cm−1 in [U(C5Me4H)3(C18O)].

Question 13.1 Suggest why, at the time of the Manhattan project in the 1940s, Gilmanattempted the synthesis of simple uranium alkyls and aryls as vehicles for isotope separation.Answer 13.1 At that time, few d-block alkyls and aryls were known. Gilman would havereasoned that by analogy with compounds of the Group IV elements like carbon and silicon,uranium(IV) alkyls and aryls might have simple molecular structures and thus be volatile.

Question 13.2 Why are simple actinide alkyls rare in view of the number of lanthanidespecies like [M{CH(SiMe3)2}3], [LnPh3(thf)3], and [LnMe6]3− (see Chapter 5)?Answer 13.2 Good question! There are several possible reasons. One is that most studieshave involved uranium, where there may be redox problems (note that CH3Li reduces Eu3+,too). Another is that the later actinides, which might be expected on size grounds to form

SPH SPH


Carbonyls 223

compounds in the +3 state similar to those of the lanthanides, are very inaccessible owingto radioactivity and non-availability of starting materials.

Question 13.3 Predict the products of the reaction of [Cp3UCl] with (a) LiNEt2, (b) LiSiPh3,(c) CH3CH2CH2CH2Li, (d) C6H5CH2MgBr. Would you expect [Cp3U(SCN)] to have U–Sor a U–N bond? Why?Answer 13.3 (a) [Cp3U(NEt2)]; (b) [Cp3U(SiPh3)]; (c) [Cp3U(CH2CH2CH2CH3)]; (d)[Cp3U(CH2C6H5)]. [Cp3U(SCN)] is N-bonded, owing to the preference of uranium for the‘hard’ donor N.

Question 13.4 Attempts to make the tert-butyl compound ThCp*2(CMe3)2 by the re-action of ThCp*2Cl2 with tert-butyllithium were unsuccessful, the hydride ThCp*2H2

{Cp*2Th(H)(µ-H)2Th(H)Cp*2} being obtained. Suggest why.Answer 13.4 The most likely answer is a steric one; attaching two bulky tert-butyl groupsto the Cp*2Th unit would result in a very congested environment.

Question 13.5 Read Section 13.4.1. Suggest a reason for the reaction of ThCp*2(CH2

CMe3)2 in forming a cyclic compound, also a reason for the reactivity of the metalla-cycles.Answer 13.5 The agostic Th....H–C interactions in the structure of ThCp*2(CH2CMe3)2

referred to earlier hint that these molecules may be congested. The cyclic compounds containa 4-membered ring; presumably the reactions relieve the strain.

Question 13.6 How would you confirm IR assignments for the bridged hydride [Cp*2Th(H)(µ-H)2Th(H)Cp*2]?Answer 13.6 Since the hydride groups are labile, isolating the deuteriated compound byexchange under a D2 atmosphere in solution should be straightforward.

Question 13.7 Suggest why the uranium compound [Cp*2U(H)(µ-H)2U(H)Cp*2] is lessstable than the thorium analogue.Answer 13.7 The +3 oxidation state is more accessible for uranium and so facilitates adecomposition pathway.

Question 13.8 Of the adducts [UCp*2(CH2SiMe3)L] (L = PMe3, py, Et2O, THF), theLewis base is most easily removed from the PMe3 adduct. Suggest why.Answer 13.8 Tertiary phosphines are ‘soft’ bases and actinides are ‘hard’ acids. This islikely to be the weakest linkage thermodynamically.

Question 13.9 The U–C and Th–C bond lengths are 2.647 and 2.701 A, respectively, inM(η8-C8H8)2(M=U, Th). In [K(diglyme)][U(η8-MeC8H7)2] the U–C bond length averages2.719 A. Explain these values.Answer 13.9 The ionic radius of Th4+ is greater than that of U4+, whilst the radius ofU3+ is greater than that of U4+.

Question 13.10 Comment upon the C–O bond length and IR data for [U(C5Me4H)3(CO)](hint, remember E = hν). The C–O distance is slightly lengthened from 1.128 A in freeCO to 1.142 A. νCO occurs at 1880 cm−1 in [U(C5Me4H)3(CO)], shifted to 1840 cm−1 in

SPH SPH



[U(C5Me4H)3(13CO)] and 1793 cm−1 in [U(C5Me4H)3(C18O)]. It occurs at 2145 cm−1 inCO itself.Answer 13.10 The C–O bond lengthens upon coordination, showing that it is gettingweaker. This can be explained in terms of uranium (5f) to CO (2π ) back-bonding, withelectrons introduced into ligand π* antibonding orbitals (thus weakening the bond). TheCO stretching frequency moves to lower frequency, also indicating bond weakening. Thisfrequency also decreases on introducing C or O atoms of higher mass, confirming thatvibrations of these atoms were indeed involved.

SPH SPH


14 Synthesis of the Transactinidesand their Chemistry


� recognize that the transactinide elements have very short half-lives;� know that studying their chemistry presents serious practical difficulties;� appreciate ways in which these elements may be synthesized and their chemistry studied;� have some knowledge of the chemistry of elements 104–108 and relate them to their

neighbours in the Periodic Table;� appreciate how the principles of the Periodic system may be used to predict the properties

of other elements yet to be made and studied.

14.1 Introduction

As noted in Section 9.2, the heaver actinides were made an ‘atom at a time’. The lastone, lawrencium, was synthesized in 1961 by bombarding a target of californium isotopes(masses 249–252) with 10B and 11B nuclei. The isotope 258Lr has a half-life of 3.8 s.

249Cf + 11B → 258 Lr + 21n

Chemical studies were not possible until 1970 with the synthesis of longer-lived 256Lr (t1/2 =26 s). The even more stable neutron-rich 262Lr with a half-life of 3.6 hours is now known. Atthe time, though, there seemed few prospects of extending the Periodic Table into anotherseries and in any case it was also not clear what chemical properties were expected for theseelements, nor how they might be studied. Experience with the actinides had shown thathalf-lives for α-decay of each element were steadily growing shorter as the atomic numberincreased, whilst the drop in half-lives for spontaneous fission was even more dramatic.Extrapolation to elements with atomic numbers around 107–108 indicated that half-liveswould be so short (t1/2 = 10−3 s) that it would be impossible to isolate further elements.

Theoreticians thought that stable heavier elements might be in prospect. The stabilityof a nucleus (based on a model of nuclear stability analogous to that of the Rutherford–Bohr model of electronic structure) is determined by the inter-nucleon forces (nucleons areprotons and neutrons), an attractive force between all nucleons and a Coulombic repulsionforce between protons, the latter becoming proportionately more important as the numberof protons increases. Extra stability is associated with filled shells of nucleons, ‘magicnumbers’; for neutrons they are 2, 8, 20, 28, 50, 82, 126, 184, and 196; and for protons theyare 2, 8, 20, 28, 50, 82, 114, and 164.


SPH SPH


226 Synthesis of the Transactinides and their Chemistry

28 50 82 126 184 196

PR

OT

ON

S

28

50

82

114

SUBMARINE

SUBMARINEMOUNTAIN

MAGICRIDGE

MAGICMOUNTAIN

N E U T R O N S

RIDGE

BETA STABILITY RIDGE

Figure 14.1Known and predicted regions of nuclear stability (reproduced with permission from G.T. Seaborg, J. Chem. Educ., 1969,46, 631).

Nuclei with closed shells of nucleons are stabilized, particularly against spontaneousfission. Nuclei where the numbers of protons and neutrons are both magic numbers (‘double-magic’ nuclei) are particularly stable, such as 4He, 16O, 40Ca, and 208Pb. After 208Pb, thenext ‘double-magic’ nucleus would be 298114. The stability of nuclei can be represented on3-D maps (Figure 14.1), with peaks corresponding to the stablest ‘magic number’ nuclei.Originally element 114 was represented by an ‘island of stability’ but new calculationssuggest relatively stable nuclei may extend towards it. In the past thirty years, elements 104–116 and 118 have all been reported and some chemistry of elements 104–108 investigated.These investigations represent one of the greatest scientific achievements.

14.2 Finding New Elements

Unsuccessful attempts were made to find elements with Z ∼ 110–114 in case they were sta-ble. Searches for element 110 (‘eka-platinum’) in platinum metal ores led to negative results,corresponding to a limit of 10−11 g per g of stable element. Similar inconclusive results wereobtained looking for element 114 (‘eka-lead’) in lead ores. Scientists therefore attemptedto synthesize these new elements, using essentially the same methods used for the lateractinides.

14.3 Synthesis of the Transactinides

Discovery of these elements has largely resulted from the researches of three groups:from Russian scientists at the Joint Institute for Nuclear Research (JINR), Dubna; Germanworkers at the Gesellschaft fur Schwerionenforschung (GSI), Darmstadt; and Americans,

SPH SPH


Synthesis of the Transactinides 227

FUSION: HOT COMPOUND

NUCLEUSDI-NUCLEUSTARGET

PROJECTILE

NEUTRON EVAPORATION

156 107

78 53

78 54

155 107

156 107

30 24

83Bi

54Cr EVAPORATION RESIDUE

FISSION FRAGMENTS

N = 126 Z = 83

262107or

209

24

Figure 14.2Synthesis of element 107 by nuclear fusion (reproduced with permission from G. Herrmann, Angew.Chem., Int. Ed. Engl., 1988, 27, 1421).

principally at the Lawrence Berkeley National Laboratory, Berkeley, California; recent con-tributions have come from Japanese workers at the RIKEN Linear Accelerator Facility. Theapproach used is to bombard a target nucleus (generally an actinide) with the positivelycharged nuclei of other elements, using a particle accelerator (cyclotron or linear acceler-ator) to supply the high energy needed to overcome the repulsion between the positivelycharged target nucleus and positive projectile nucleus, as already discussed in Section 9.2.In this method the atoms are produced one at a time, obviating ‘normal’ chemical experi-ments on bulk samples. The chance of a collision producing a combined nucleus is at bestabout one in a billion collisions.

Moreover, since the new atomic nucleus produced is in an excited nuclear state, it has anexcess of energy to get rid of, whether by emission of neutrons and γ -rays, or by fission(Figure 14.2) (If the energy of the bombarding particles is too low, they do not possessenough energy to overcome the Coulombic repulsion between them.)

For elements 104 and 105, competing claims of first discovery have been made by theAmerican and Russian teams. In 1969 workers from Berkeley reported the synthesis of257–259104:

249Cf + 12C → 257104 + 4 1n; 249Cf + 13C → 259104 + 3 1n248Cm + 16O → 258104 + 6 1n

Almost simultaneously an alternative route was reported by the Dubna workers:

242Pu + 22Ne → 260104 + 41n; 242Pu + 22Ne → 259104 + 3 1n

Since different isotopes of the transactinides have frequently been reported from differentsynthetic routes, it has been very difficult to decide who ‘discovered’ a new element.

Element 105 was reported by the Berkeley group in 1970 and the Dubna group in 1971.The Berkeley team used

249Cf + 15N → 260105 + 41n

whilst the Dubna group employed

243Am + 22Ne → 260105 + 5 1n

SPH SPH



The original (1974) synthesis of element 106 by the Berkeley group was repeated thereby a different team in 1993:

249Cf + 18O → 263106 + 41n

whilst heavier isotopes of mass 265 and 266 with half-lives of 7 s. and 21 s, respectively,suitable for chemical studies, are now known.

248Cm + 22Ne → 265106 + 51n; 248Cm + 22Ne → 266106 + 41n

Elements 107–112 were initially synthesized by the Darmstadt group, though subse-quently other isotopes have been made by the other groups or detected in the decay chainsof heavier nuclei. Their syntheses used rather heavier projectile atoms and concomitantlylighter targets; this is sometimes called a ‘cold-fusion’ approach (not to be confused withthe notorious room-temperature fusion reactions reported in 1989 as a route to ‘limitlessenergy’). This method uses projectile particles with the minimum kinetic energy to justresult in fusion so that the product nucleus has little excessive energy to get rid of, so thatnot only is the likelihood of fission minimized, but very few neutrons ‘evaporate’ fromthe ‘hot’ nucleus, so that the mass number of the resulting product is as high as possible.Another significant point is that the Pb and Bi targets have magic- or near-magic-numberstructures, as do the nickel projectile nuclei used in the syntheses of elements 110 and 111,so that a large part of the projectile energy is consumed in breaking into the filled shell andthis also reduces the possibility of the new nucleus undergoing fission.

209Bi + 54Cr → 262107 + 1n; 208Pb + 58Fe → 265108 + 1n209Bi + 58Fe → 265109 + 21n; 208Pb + 62Ni → 269110 + 1n208Pb + 64Ni → 271110 + 1n; 209Bi + 64Ni → 272111 + 1n208Pb + 70Zn → 277112 + 1n

A different approach to element 107 has led to an isotope with a half-life of 14 s, used inchemical studies; similarly, synthesis of 269108 with a half-life of around 11s has permittedthe first chemical investigations of element 108.

249Bk + 22Ne → 267107 + 41n; 248Cm + 26Mg → 269108 + 51n

In 1999–2000, syntheses of elements 114, 116, and 118 were first reported, but the resultsfor experiment 118 have proved to be irreproducible, so it cannot yet be claimed to havebeen made.

Bombardment of 244Pu with neutron-rich 48Ca for several weeks led to reports of oneatom of element 114. The atom had a lifetime of about 30 s before decaying to form anatom of element 112:

244Pu + 48Ca → 289114 + 31n; 289114 → 285112 + 4He

Particularly long-lived atoms detected in the decay chain were identified as isotopes ofelements 112 and 108 with respective lifetimes of around 15 minutes. Other isotopes 287114and 288114 have also been reported.

A similar route, bombardment of 248Cm with 48Ca, was used to make element 116 (Uuh);it alpha-decayed after 47 ms to a known isotope of element 114:

248Cm + 48Ca → 292116 + 41n; 292116 → 288114 + 4He

Isotopes 290116 and 291116 have also been reported.

SPH SPH


Naming the Transactinides 229

An alternative route was followed by workers at LBNL in California, who used a ‘coldfusion’ approach with a heavier projectile nucleus (krypton) and a lighter target (lead);bombardment over an eleven-day period afforded three atoms having a lifetime of lessthan a millisecond, believed to be element 118, though these claims must be regarded asunsubstantiated.

208Pb + 86Kr → 293118 + 1n

Most recently, in February 2004, experiments were reported giving serious evidence forelements 113 and 115, again synthesized using 48Ca as the projectile:

243Am + 48Ca → 287115 + 41n (3 atoms)243Am + 48Ca → 288115 + 31n (1 atom)

These atoms all decayed (with lifetimes in the range 19–280 milliseconds) by α-decay toafford what are believed to be atoms of element 113:

287115 → 283113 + 4He; 288115 → 284113 + 4He

These atoms of element 113 have longer lifetimes, one in excess of a second.An alternative route for synthesis of an atom of element 113 was reported by a team of

Japanese (largely) and Chinese workers in October 2004:

209Bi + 70Zn → 278113 + 1n (1 atom)

This decayed with a lifetime of 344 ms to an isotope of element 111.

278113 → 274111 + 4He; 274111 → 270109 + 4He

14.4 Naming the Transactinides

The two groups who claim to have synthesized elements 104 and 105 gave them both aname. The International Union of Pure and Applied Chemistry therefore suggested a systemof surrogate names for these elements until definite identification could be credited; thisassigned each element a name based on its atomic number, using the ‘code’ 0 = nil, 1 = un,2 = bi, 3 = tri, 4 = quad, 5 = pent, 6 = hex, 7 = sept, 8 = oct, and 9 = enn. Thus element104 is unnilquadium.

In 1997, the International Union of Pure and Applied Chemistry decided names forelements 104–109, which have generally met with acceptance; they have recently decidednames for 110 (2003) and for 111 (2004).

No decisions have yet been made about the heavier elements.

Table 14.1

Atomic number Name Symbol

104 Rutherfordium Rf105 Dubnium Db106 Seaborgium Sg107 Bohrium Bh108 Hassium Hs109 Meitnerium Mt110 Darmstadtium Ds111 Roentgenium Rg

SPH SPH



14.5 Predicting Electronic Arrangements

The 5f orbitals are gradually filled as the actinide series is crossed and there is generalagreement that the 6d orbitals will be filled next, followed by 7p orbitals. Table 14.2 listspredicted electron configurations.

Table 14.2 Election configurations of elements 104–121a

Element 6d 7s 7p 1/2 7p 3/2 8s 7d (or 8p 1/2)

104 2 2105 3 2106 4 2107 5 2108 6 2109 7 2110 8 2111 9 2112 10 2113 10 2 1114 10 2 2115 10 2 2 1116 10 2 2 2117 10 2 2 3118 10 2 2 4119 10 2 2 4 1120 10 2 2 4 2121 10 2 2 4 2 1

a Electron configurations are given outside the core [Rn] 5f 14.

14.6 Identifying the Elements

Classical chemical routes are impractical with ultrasmall quantities of very radioactive,short-lived atoms. Even the most stable transactinides have such short half-lives (Table14.3) that they had to be rapidly removed from the accelerator target for chemical study.This is achieved by the recoil of the new atom from the target and its transport by an aerosolin a current of helium to a collection site. Figure 14.3 shows the system used to study thechlorination of elements 104 and 105, using an MoO3 aerosol to transport the atoms prior tohalogenation. The volatile halides formed are transported by the helium along the latter partof the tube, which acts as an isothermal gas chromatography column; on leaving the quartztube they are attached to new aerosols and detected either with a rotating wheel system oron a moving computer tape.

The millisecond half-lives of the first isotopes of elements 107–112 meant that therewas no chance of carrying out chemical separations, so the Darmstadt group have used a‘velocity filter’ which relies on electric and magnetic fields to separate the excess of fast-moving ‘projectile’ nuclei from slow-moving products recoiling from the target so that thereaction products can be detected (Figure 14.4). The beams of particles are focussed withmagnetic lenses and then separated by electric and magnetic fields before colliding withsolid-state detectors.

SPH SPH


Identifying the Elements 231

Table 14.3 Half-lives and decay modes for the isotopes of elements 104–116 and 118

Element Half-life Decay mode

104253 0.05 ms SF254 0.5 ms SF255 1.4 s SF, α256 7 ms α, SF257 4.8 s α, SF258 13 ms SF, α?259 3.0 s α, SF260 20 ms SF261 78 s α, SF262 47 ms SF?263 10 m α, SF

105255 1.6 s α, SF256 2.6 s EC, α257 1.3 s α, SF258 20 s EC, α259 α260 1.5 s α, SF261 1.8 s α, SF262 34 s α, SF263 27 s SF, α265 7.4 s α266 21 s α

106258 2.9 ms SF259 0.9 s α, SF260 3.6 ms α, SF261 0.23 s α, SF?263 0.8 s α, SF265 16 s α266 21 s α

107260 α261 12 ms α, SF?262 8 ms/0.1 s α264 0.44 s α267 17 s α

108263 <0.1s α264 0.08 ms α, SF265 1.8 ms α267 33 ms α269 11 s α270 3.6 s α273 1.2 s α277 11.4 m α

109266 3.4 ms α268 70 ms α

Continued

SPH SPH


Table 14.3 Continued

Element Half-life Decay mode

110267 4 µs α269 0.17 ms α271 1.1 ms α273 0.1 s/0.1 ms α277280 7.4 s SF281 1.1 m α

111272 1.5 ms α

112277 242 µs α281283 3 m SF284 9.7 s α285 10.7 m α

113283 ∼1 s α284 ∼0.4 s α

114285 <1 ms α287 5 s α288 1.9 s α289 ∼21 s α

115287 ∼46 ms α288 20–200 ms α

116289 47 ms α

118293 α

Figure 14.3Apparatus for the study of the volatile halides of elements 104 and 105 (reproduced with permission from J.V. Kratz, J.Alloys Compd., 1994, 213–214, 22).

232

SPH SPH


Predicting Chemistry of the Transactinides 233

Figure 14.4Velocity filter SHIP to separate transactinide fusion products from the projectile beam (reproduced with permission fromG. Herrmann, Angew. Chem., Int. Ed. Engl., 1988, 27, 1422).

Detection of these elements relies on studying their radioactive decay, which is usuallyeither α-emission or spontaneous fission. A time-correlation process is used; in this, a solid-state detector monitors both the time and position of arrival of fusion products. Subsequentdecay events at this position give not just the decay information of the atom (half-life,α-particle energy) but also the corresponding information for its decay products, which arerecognizable and thus ‘known’ nuclei. This therefore gives a history of stepwise decay ofthe initial product.

Thus reaction of 209Bi with 54Cr led to very short-lived nuclei (t1/2 = 4.7 ms) emittingα-particles with an energy of 10.38 MeV. This α-emitter correlated with a decay chaincomposed of three successive α-emitters, identified as 258105 (4.4 s; α-particles of 9.17MeV); 254Lr (13 s; α-particles of 8.46 MeV); and 250Md (52 s; α-particles of 7.75 MeV).The product of the fusion reaction was thus 262107.

209Bi + 54Cr → 262Bh + 1n262107 → 258Db + 4He258105 → 254Lr + 4He

254Lr → 250Md + 4He

The limiting factor in detecting the reaction product is thus the time of flight of the productfrom the target to the detector, which is of the order of microseconds.

14.7 Predicting Chemistry of the Transactinides

Elements from 104 to 112 are believed to have electron configurations resulting from fillingthe 6d orbitals, so they are predicted to be transition metals. For the elements with Z > 112,it is assumed that the 7p orbitals are filled first; elements 119 and 120 involve filling the8s orbital; whilst from 121 to 153, one electron is first placed in the 7d (or 8p) orbital

SPH SPH



then the 5g and 6f orbitals. Relativistic effects (Section 9.7) can cause the elements to haveproperties differing from those predicted by simple extrapolation.

14.8 What is known about the Chemistry of the Transactinides

14.8.1 Element 104

The aqua ion of Rf is eluted as the α-hydroxybutyrate complex from cation-exchange resinsmuch earlier than are trivalent actinides, because it is more weakly adsorbed by the resin,its behaviour resembling that of Zr and Hf. This supports the assignment of a (+4) oxida-tion state. A coordination number of 6 for Rf 4+(aq) has been suggested, with a predictedionic radius of 78 pm for Rf 4+. It is extracted from concentrated HCl as anionic chloridecomplexes, unlike trivalent actinides and group I and II metals; these complexes, possiblyinvolving Rf Cl62−, are eluted from anion-exchange resins with zirconium and hafnium butlater than actinides. Like Zr and Hf, however, it forms stable anionic fluoride complexes, anda complex ion RfF6

2− has been proposed. The volatility of RfCl4 (condensing ∼220 ◦C) issimilar to that of ZrCl4 but greater than that of HfCl4; it is much greater than that of actinidetetrachlorides. RfBr4 is more volatile than Hf Br4 (though the bromide is less volatile thanthe chloride).

14.8.2 Element 105

The aqua ion of Db is readily hydrolysed, even in strong HNO3, and resembles those ofNbV, TaV and PaV rather; the +4 ions of the Group IV elements don’t do this, so this isan argument in favour of the +5 oxidation state in aqueous solution. Similarly, elution ofDb as the α-hydroxybutyrate complex from cation-exchange resins shows it to be elutedrapidly, like NbV, TaV, and PaV but unlike Zr4+ and Eu3+ ions, which are strongly retainedon the column. Study of complex formation in HCl indicates resemblances to NbV andPaV rather than TaV. It forms anionic fluoride complexes which, like those of Nb, Ta, andPa, are strongly adsorbed on anion-exchange resins. Study of the volatility of the halides(assumed to be pentahalides) indicates that DbBr5 is less volatile than NbBr5 and TaBr5

(in contradiction of recent theoretical predictions) and Rf Br4. The chloride appears to bemore volatile than the bromide (at present, it still remains to be confirmed that it was thebinary halides studied in these experiments rather than oxyhalides, as well as confirmingthe oxidation state of the element).

14.8.3 Element 106

A comparative study of the reaction of 263Sg, Mo, and W with SOCl2 in air suggests thatSgCl2O2 is formed, with similar volatility to the Mo and W homologues. Investigationsof Sg in aqueous solution indicate +6 to be the most stable oxidation state and that itforms neutral or ionic oxo and oxohalide compounds, species possibly including [SgO4]2−,[SgO3F]−, or [SgO2F4]2−. Chromatographic study of solutions in very dilute HF indicatesthe formation of SgO4

2−, analogous to MoO42− and WO4

2−. So far, ‘seaborgium’ behavesas a 6d transition metal, similar to Mo and W, with no evidence for deviations due torelativistic effects and little resemblance to uranium.

SPH SPH


What is known about the Chemistry of the Transactinides 235

14.8.4 Element 107

In 2000, chemists synthesized atoms of 267Bh (t1/2 ∼ 17 s), using the reaction 22Ne (249Bk,

4n) 267Bh. Atoms recoiling from the back of the berkelium target were absorbed onto carbonparticles suspended in a flow of helium gas and transported to a quartz wool trap in an ovenat 1000 ◦C. There the carbon and the nuclear reaction products were reacted with a mixtureof HCl and O2 gases. Some six atoms of bohrium were detected as a volatile oxychloride,believed to be BhO3Cl, by analogy with Tc and Re.

14.8.5 Element 108

In the only studies so far reported (2005), hassium atoms were oxidized to a very volatileoxide (thought to be HsO4 by analogy with Ru and Os – its absorption enthalpy is comparablewith that of OsO4), and deposited on an NaOH surface. Six correlated α-decay chains of Hswere observed and attributed to Na2[HsO4(OH)2], sodium hassate(VIII), by analogy withosmium, the 5d homologue of Hs.

Figure 14.5‘Future’ Periodic Table (reproduced with permission from G.T. Seaborg, J. Chem. Soc., Dalton Trans., 1996, 3904).

SPH SPH



14.9 And the Future?

Already many more elements have been synthesized than could be predicted in the early1960s. Higher neutron-flux reactors could make the necessary amounts of the heaviertransuranium elements needed for synthesis of the elements beyond 112. Much will de-pend upon whether the more neutron-rich isotopes with longer half-lives, essential for anystudy of chemical properties, can be made. The only safe prediction is the unpredictabilityof this area.

Question 14.1 Study the electron configurations in Table 14.2, and use your knowledge ofthe properties of lighter elements to predict properties of the following elements and theircompounds:(a) Element 104 and its compounds(b) Element 106 and its compounds(c) Element 111 and its compounds(d) Element 114 and its compounds(e) Element 118 and its compoundsAnswer (a) Element 104 falls below Hf in the Periodic Table. It should have the electronconfiguration [Rn] 5f 14 6d2 7s2. Chemically it should resemble zirconium and hafnium,with a maximum oxidation state of +4, which is likely to be the most stable, though some+3 compounds should exist. It should form an oxide MO2 and volatile halides, MX4.Answer (b) Element 106 falls below W in the Periodic Table. It should have the electronconfiguration [Rn] 5f 146d4 7s2. Chemically it should resemble tungsten. It could have oxi-dation states up to +6, though some have suggested the +4 state might be most likely forthe aqua ion. It ought to form some volatile halides, including MF6. It has been suggestedthat element 106 might form a stable M(CO)6, as do Mo and W.Answer (c) Element 111 falls below Au in the Periodic Table. It should have the electronconfiguration [Rn] 5f 146d9 7s2. Chemically it should resemble gold and be a rather unreac-tive metal. Its most stable oxidation state is likely to be +3, though there may be a fluorideMF5. It may form an ion M−, analogous to Au−. The oxides may be too unstable to existat room temperature.Answer (d) Element 114 falls below Pb in the Periodic Table. It should have the electronconfiguration [Rn] 5f 146d10 7s2 7p2. Chemically it should resemble lead. Its most stableoxidation state is likely to be +2, and compounds in the +4 state may be too unstable toexist at room temperature. It should form several compounds, including an oxide MO andseveral halides MX2; by analogy with lead, the chloride would be insoluble but a nitratewould be soluble. Element 114 could be a rather noble metal (though it has been suggestedthat the element might be a liquid or even a gas at room temperature!).Answer (e) Element 118 falls below Rn in the Periodic Table. It should have the electronconfiguration [Rn] 5f 146d10 7s2 7p6. Chemically it should resemble radon, but should bethe most reactive noble gas (though simple extrapolation of trends down the Group suggestsit could be a liquid element at room temperature). It could form compounds in oxidationstates including +2 and +4, with fluorides like MF2 and MF4 as well as a relatively stablechloride MCl2; because element 118 should be less electronegative than the other noblegases, these compounds might be rather ionic and involatile.

SPH SPH


Bibliography

J. Emsley, The Elements, OUP, 1997; Nature’s Building Blocks, OUP, 2001.D.R. Lide (ed.), CRC Handbook of Chemistry and Physics, 85th edition, Central Rubber Company

Press, 2004.

Acronyms for General Sources

AIP N.M. Edelstein (ed.), Actinides in Perspective, Pergamon, 1982.ACE J.J. Katz, G.T. Seaborg and L.R. Morss (eds), The Chemistry of the Actinide Elements, 2nd

edition, Chapman and Hall, 1986.APU D. Hoffman (ed.), Advances in Plutonium Chemistry 1967–2000, American Nuclear Society,

2002.CCC-1 G. Wilkinson, R.D. Gillard and J.A. McCleverty (eds), Comprehensive Coordination Chem-

istry, 1st edition, Pergamon, Oxford, 1987.CCC-2 J.A. McCleverty and T.J. Meyer (eds), Comprehensive Coordination Chemistry II, Elsevier,

Amsterdam, 2004.CIC J.C. Bailar Jr., H.J. Emeleus, R.S. Nyholm and A.F. Trotman-Dickenson (eds), Comprehensive

Inorganic Chemistry, Pergamon, Oxford, 1973.COMC-1 G. Wilkinson, F.G.A. Stone and E. Abel (eds), Comprehensive Organometallic Chemistry,

Pergamon, Oxford, 1982.COMC-2 E.W. Abel, F.G.A. Stone and G. Wilkinson (eds), Comprehensive Organometallic

Chemistry-2, Pergamon, Oxford, 1995.EIC-1 R.B. King (ed.), Encyclopedia of Inorganic Chemistry, 1st edition, Wiley, Chichester, 1994.EIC-2 R.B. King (ed.), Encyclopedia of Inorganic Chemistry, 2nd edition, Wiley, Chichester, 2005.GMEL Gmelins Handbuch der Anorganischer Chemie.HAC A.J. Freeman and G.H. Lander (eds), Handbook on the Physics and Chemistry of the Actinides,

North Holland, 1984 onwards, Vols 1–5.HRE K.A. Gschneider, LeRoy Eyring, et al. (eds), Handbook on the Physics and Chemistry of Rare

Earths, North Holland, 1978 onwards, Vols 1–33 (at present).IAR K.A. Gschneider (ed.), Industrial Applications of Rare Earth Elements, ACS Symposium

Series, 1981.LAS N.M. Edelstein (ed.) , Lanthanide and Actinide Chemistry and Spectroscopy, vol. 131. ACS

Symposium Series, 1980.LPL J.-C.G. Bunzli and G.R. Choppin (eds), Lanthanide Probes in Life, Chemistry and Earth

Sciences, Elsevier, 1989.MTP 1 and MTP2 K.W. Bagnall (ed.), MTP International Review of Science, Inorganic Chem-

istry, Series 1 (1972) vol. 7; Series 2 (1975) vol. 7.


SPH SPH


238 Bibliography

PIC S.F.A. Kettle, Physical Inorganic Chemistry: A Coordination Chemistry Approach, Spektrum,1996.

SPL S.P. Sinha (ed.) Systematics and Properties of the Lanthanides, ACS Symposium Series, 1981.


Further Reading

D.M. Yost, H. Russell, Jr and C.S. Garner, The Rare Earths and Their Compounds, Wiley, 1947.T. Moeller, CIC-1, 4, 1.N.E. Topp, The Chemistry of the Rare Earth Elements, Elsevier, 1965.G.R. Choppin, LPL, 1.B.T. Kilbourn, A Lanthanide Lanthology, Molycorp, 1993 (Parts I and II).B.T. Kilbourn, Cerium. A Guide to its Role in Chemical Technology, Molycorp, 1992.T. Moeller, J. Chem. Educ., 1970, 47, 417 (principles).C.H. Evans (ed), Episodes from the History of the Rare Earth Elements, Kluwer, Dordrecht, 1996.F. Szabadvary, HRE, 1988, 11, 33 (discovery and separation of lanthanides).R.G. Bautista, HRE, 1995, 21, 1 (separation chemistry).K.L. Nash, HRE, 1991, 18, 197 (separation chemistry of trivalent lanthanides and actinides).H.E. Kremers, J. Chem. Educ., 1985, 62, 445 (extraction).G.W. de Vore, LPL, 321 (geochemistry).J.-C. Duchesne, SPL, 543 (geochemistry).P. Moller, SPL, 561 (geochemistry).S.R. Taylor and M.M. McLennan, HRE, 11, 485 (geochemistry).L.A. Haskin and T.P. Paster, HRE, 3, 1 (geochemistry).R.H. Byrne and E.R. Sholkovitz, HRE, 1996, 23, 497 (marine chemistry and geochemistry).

General References

B.H. Ketelle and G.E. Boyd, J. Am. Chem. Soc., 1947, 69, 2800 (chromatography).F.H. Spedding, E.I. Fulmer, J.E. Powell, and J.A. Butler, J. Am. Chem. Soc., 1950, 72, 2354

(chromatography).W.B. Jensen, J. Chem. Educ., 1982, 59, 634 (positions of La and Lu in the Periodic Table).


Further Reading

S.F.A. Kettle, PIC, 240 (f orbitals).G.T. Seaborg and W.D. Loveland, The Elements beyond Uranium, Wiley, 1990, pp. 71–78 (f electrons).D.R. Lloyd, J. Chem. Educ., 1986, 63, 502 (lanthanide contraction).D.A. Johnson, J. Chem. Educ., 1980, 57, 475 (reactivity and 4f electrons).L.J. Nugent, J. Inorg. Nucl. Chem., 1975, 37, 1767; MTP2, 7, 195 (redox potentials, oxidation states).L.R. Morss, HRE, 1991, 18, 239 (redox potentials).D.A. Johnson, Some Thermodynamic Aspects of Inorganic Chemistry, 2nd edition, CUP, Cambridge,

1982, pp. 50–54, 158–168; J. Chem. Educ., 1986, 63, 228 (energetics).D.W. Smith, Inorganic Substances, CUP, Cambridge, 1990, pp. 145–149, 163–167 (energetics).

SPH SPH


Bibliography 239

D.A. Johnson., Adv. Inorg. Chem. Radiochem., 1977, 20, 1 (unusual oxidation states).F.A. Hart, CCC-1, 3, 1109, 1113 (unusual oxidation states).T. Moeller, CIC, 4, 75, 97 (unusual oxidation states).

General References

D.A. Johnson, J. Chem. Educ., 1980, 57, 475 (ionic radii of Lanthanide ions).D.W. Smith, J. Chem. Educ., 1986, 63, 228 (enthalpies of formation).H.G. Friedman, G.R. Choppin, and D.G. Feverbacher, J. Chem. Educ., 1964, 41, 354 (f orbitals).


Further Reading

B.J. Beaudry and K.A. Gschneider Jr, HRE, 1, 173 (elements).H.F. Lineberger, T.K. McCluhan, L.A. Luyckx and K.E. Davies, IAR, 3, 19, 167 (applications).V. Paul-Boncour, L. Hilaire and A. Percheron-Guegan, HRE, 2000, 29, 5 (metals and alloys in

catalysis).GMEL: C1 (oxides), C2 (hydroxides), C3, C4a/b, C6 (halides), C7 (sulfides), C11a/b (borides).L. Eyring, HRE, 3, 337 (oxides).R.G. Haire and L. Eyring, HRE, 1991, 18, 413 (oxides).G. Adachi and N. Imanaka, Chem. Rev., 1998, 98, 1479 (oxides).J. Burgess and J. Kijorski, Adv. Inorg. Chem. Radiochem., 1981, 24, 51 (halides).J.M. Haschke, HRE, 4, 89 (halides).H.A. Eick , HRE, 1991, 18, 365 (lanthanide and actinide halides).G. Meyer and M. Wickleder, HRE, 2000, 28, 53 (simple and complex halides).A. Simon, Hj. Mattausch, G.J. Miller, W. Bauhofer and R.K. Kremer, HRE, 1991, 15, 191 (metal-rich

halides).J.G.-y. Adachi, N. Imanaka and Z. Fuzhong, HRE, 1991, 15, 61 (carbides).E.L. Huston and J.J. Sheridan III, IAR, 223 (hydrides).G.G. Libowitz and A.J. Maeland, HRE, 3, 299 (hydrides).J.W. Ward and J.M. Haschke, HRE, 1991, 18, 293 (hydrides).Yu. Kuz’ma and S. Chykhrij, HRE, 1996, 23, 285 (phosphides).I.G. Vasilyeva, HRE, 2001, 32, 567 (polysulfides).K. Mitchell and J.A. Ibers, Chem. Rev., 2002, 102, 1929 (rare-earth and transition-metal chalco-

genides).

General References

S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991.R.B. King (ed.), Encyclopedia of Inorganic Chemistry, 1st edition, Wiley, Chichester, 1994.

Superconductors

Further Reading

C.N.R. Rao, (ed), Chemistry of Oxide Superconductors, Blackwell, Oxford, 1988.M.B. Maple, M.C. de Andrade, J. Herrmann, R.P. Dickey, N.R. Dilley and S. Han, J. Alloys Compd.,

1997, 250, 585 (superconductors).

SPH SPH


240 Bibliography

Handbook on the Physics and Chemistry of Rare Earths, North Holland, 2000–2001, Volumes 30 and31, contains several reviews on high-temperature superconductors; the most useful of these are thefollowing:

� M.B. Maple, HRE, 2000, 30, 1 (superconductivity in layered cuprates);� B. Raveau, C. Michel and M. Hervieu, HRE, 2000, 30, 31 (crystal chemistry of superconducting

rare-earth cuprates);� W.E. Pickett and I.I. Mazin, HRE, 2000, 30, 453 (electronic theory and physical properties of

RBa2Cu3O7 compounds);� K. Kobayashi and S. Hirosawa, HRE, 2001, 32, 515 (permanent magnets).

General References

M.K. Wu, J.R. Ashburn, C.J. Torng, P.H. Hor, R.L. Meng, L. Gao, Z.J. Huang, Q. Wang and C.W.Chu, Phys. Rev. Lett., 1987, 58, 908 (YBa2Cu3O7−8).

K.A. Bednorz and J.G. Muller, Science, 1987, 237, 1133; Angew. Chem. Int. Ed. Engl., 1988, 27, 735(superconducting oxides).

4 Coordination Chemistry of the Lanthanides

Further Reading

F.A. Hart, CCC-1, 3, 1087 (complexes).S.A. Cotton, CCC-2, 3, 93; EIC-2, p. 4838 (complexes).L. Niinsto and M. Leskala, HRE, 8, 203; 9, 91 (complexes).L.C. Thompson, HRE, 3, 209 (complexes).J.F. Desreux, LPL, 43 (complexes).GMEL, D1–D5 (complexes).S.A. Cotton, CR Chimie, 2005, 8, 129 (establishing coordination numbers for the lanthanides in simple

complexes).G.R. Choppin, J. Alloys Compd., 1997, 249, 1 (factors involved in lanthanide complexation).T. Moeller, CIC, 4, 28; MTP1, 285 (stability constants).A.E. Martell and R.M. Smith, Critical Stability Constants, Plenum, 1974 onwards, vols 1–6.H.C. Aspinall, Chem. Rev., 2002, 102, 1807 (chiral lanthanide complexes).E.N. Rizkalla and G.R. Choppin, HRE, 1991, 15, 393; HRE, 1994, 18, 529 (hydration and hydrolysis

of lanthanides).N. Marques, A. Sella and J. Takats, Chem. Rev., 2002, 102, 2137 ( pyrazolylborate complexes of the

lanthanides).G.R. Choppin, J. Alloys Compd., 1993, 192, 256 (aminopolycarboxylate complexes).G. Bombieri and R. Artali, J. Alloys Compd., 2002, 344, 9 (DOTA complexes).J.C.G. Bunzli, HRE, 9, 321; LPL, 219 (macrocycle complexes).A.G. Coutsolelos, NATO ASI Ser, Ser. C., 1995, 459, 267 (porphyrin complexes).D.K.P. Ng and J. Jiang, Chem. Soc. Rev., 1997, 26, 433 (porphyrin complexes).L.M. Vallarino, HRE, 1988, 15, 4433 (macrocycle complexes from template syntheses).R.C. Mehrotra, A. Singh and U.M. Tripathi, Chem. Rev., 1991, 91, 1287 (lanthanide alkoxides).M.N. Bochkarev, L.N. Zakharov and G.S. Kalinina, Organoderivatives of Rare Earth Elements,

Kluwer, Dordrecht, 1995 (amides, alkoxides, thiolates).D.C. Bradley, R.C. Mehrotra, I.P. Rothwell and A. Singh, Alkoxo and Aryloxo Derivatives of Metals,

Academic Press, San. Diego, 2001.

SPH SPH


Bibliography 241

General References

G.B. Deacon, T. Feng, P.C. Junk, B.W. Skelton, A.N. Sobolev, and A.H. White, Aurt. J. Chem., 1998,51, 75 (lanthanide chloride complexes of THF).

G.B. Deacon et al., J. Chem. Soc., Dalton Trans., 2000, 961 (alkoxides).X.-Z. Feng, A.-L. Guo, Y.-T. Xu, X.-F. Li and P.-N. Sun, Polyhedron, 1987, 6, 1041 (coordination

numbers).J. Marcalo and A. Pires de Matos, Polyhedron, 1989, 8, 2431 (coordination numbers).B.P. Hay, Inorg. Chem., 1991, 30, 2881 (nonaqualanthanide ion structure).A. Habenschuss and F.H. Spedding, Cryst. Struct. Commun., 1982, 7, 538 (crystal structure Ln cation).A. Zalkin, D.H. Templeton, and D.G. Karraker, Inorg. Chem., 1969, 8, 2680 (structure of Ho complex).T. Phillips, D.E. Sands, and W.F. Wagner, Inorg. Chem., 1968, 7, 2299 (structure of La complex).J.D.J. Backer-Dirks, J.E. Cooke, A.M.R. Galas, J.S. Ghotra, C.J. Gray, F.A. Hart, and M.B. Hursthouse,

J. Chem. Soc., Dalton Trans., 1980, 2191 (structure of a La crown ether complex).S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991, p. 64 (structure of La EDTA complex).M. Frechette, I.R. Butler, R. Hynes, and C. Detellier, Inorg. Chem, 1992, 31, 1650 (structure of La

phen complex).G.H. Frost, F.A. Hart, C. Heath, and M.B. Hursthouse, Chem. Commun., 1969, 1421 (structure of Eu

terpy complex).O. Poncelet, O.W.J. Sartain, L.G. Hubert-Pfalzgraf, K. Folting, and K.G. Caulton, Inorg. Chem., 1989,

28, 264 [structure of Y5O(O Pri)13].D.L. Clark, J.C. Gordon, P.J. Hay, R.L. Martin, and R. Poli, Organometallics, 2002, 21, 5000

(theoretical calculation).L. Perrin, L. Maron, O. Eisenstein, and M.F. Lappert, New J. Chem., 2002, 27, 121 (theoretical

calculation).

5 Electronic/Magnetic Properties of the Lanthanides

Magnetism

Further Reading

F.A. Hart, CCC, 3, 1109.T. Moeller, CIC, 4, 9.J. Rossat-Mignod, SPL, 255.J.J. Rhyne, HRE, 1988, 11, 293.S.F.A. Kettle, PIC, 265.

Electronic Spectra

Further Reading

F.A. Hart, CCC-1, 3, 1105.T. Moeller, CIC, 4, 9; MTP1, 7, 275.W.T. Carnall, J.V Beitz, H. Crosswhite, K. Rajnak and J.B. Mann, SPL, 389; HRE, 3, 171.C.A. Morrison and R.P. Leavitt, HRE, 5, 461.J.W. O’Laughlin, HRE, 4, 341.B.R. Judd, HRE, 11, 81 (atomic theory and optical spectroscopy).B.R. Judd, LAS, 267.J.V. Beitz, HRE, 1991, 18, 159 (absorption and luminescence spectra).

SPH SPH


242 Bibliography

S.F.A. Kettle, PIC, 257 (f–f transitions).C. Gorller-Walrand and K. Binnemans, HRE, 1996, 23, 121 (crystal-field parameters).C. Gorller-Walrand and K. Binnemans, HRE, 1998, 25, 101 (intensities of f–f transitions).N. Sabattini, M. Guardigli and I. Manet, HRE, 1996, 23, 69 (antenna effect in encapsulated complexes).G.E. Buono-core and H.LiB. Marciniak, Coord. Chem. Rev., 1990, 99, 55 (quenching of excited states

by lanthanide ions and chelates in solution).GMEL, E1, 1993 (optical spectra of praseodymium compounds).GMEL, E2a, 1997 (optical spectra of neodymium compounds).

General References

D.G. Karraker, Inorg. Chem., 1968, 7, 473 (hypersensitivity).S. Hufner, SPL, 313.

Luminescence spectra

Further Reading

C. Fouassier, EIC-1, 1984; EIC-2, 2730 (luminescence).K.B. Yatmirskii and N.H. Davidenko, Coord. Chem. Rev., 1979, 27, 223 (spectra).J.H. Forsberg, Coord. Chem. Rev., 1973, 10, 195 (spectra).S.F.A. Kettle, PIC, 263 (luminescence).D. Parker, Coord. Chem. Rev., 2000, 205, 109 (luminescent sensors).Y. Korovin and N. Rusakova, Rev. Inorg. Chem., 2001, 21, 299 (IR 4f luminescence).I. Billard, HRE, 2003, 33, 465 (time-resolved, laser-induced emission spectroscopy).D. Parker and J.A.G. Williams, J. Chem. Soc., Dalton Trans., 1996, 3613 (absorption and lumines-

cence spectra, NMR).D. Parker, R.S. Dickins, H. Puschmann, C. Crossland and J.A.K. Howard, Chem. Rev., 2002, 102,

1977 (absorption and luminescence spectra, NMR).D. Parker, Chem. Soc. Rev., 2004, 33, 156 (absorption and luminescence spectra, NMR).T. Gunnlaugsson and J.P. Leonard, Chem. Commun., 2005, 3114 (luminescent sensors).G.R. Choppin and Z.M. Wang, Inorg. Chem., 1997, 36, 249 (7F0 → 5D0 transition).S.T. Frey and W.D. Horrocks, Inorg. Chim. Acta., 1995, 229, 383 (7F0 → 5D0 transition).G-y. Adachi and N. Inamaka, HRE, 1995, 21, 179 (chemical sensors).G. Blasse, Chem. Mater., 1994, 6, 1465 (scintillator materials).

General References

T. Gunnlaugsson, J.P. Leonard, K. Senechal, and A.R. Harte, J. Am. Chem. Sec., 2003, 125, 12062(luminescence spectra).

T. Gunnlaugsson, J.P. Leonard, K. Senechal, and A.R. Harte, Chem. Commun., 2004, 782 (lumines-cence spectra).

J.-C.G. Bunzli, B. Klein, G. Chapuis, and K.J. Schenk, Inorg. Chem., 1982, 21, 808 (luminescencespectra).

T. Yamada, S. Shinoda, and H. Tsukube, Chem. Commun., 2002, 218 (luminescent sensors).T. Gunnlaugsson, D.A. MacDonail, and D. Parker, Chem. Commun., 2000, 93 (luminescent sensors).D. Parker and J.A.G. Williams, Chem. Commun., 1998, 245 (luminescent sensors).

SPH SPH


Bibliography 243

D. Parker, K. Senanaayake, and J.A.G. Williams, Chem. Commun., 1997, 1777 (luminescent sensors).B. Blasse and A. Bril, J. Chem. Phys., 1966, 45, 3327 (luminescence).D.A. Durham, G.H. Frost, and F.A. Hart, J. Inorg. Nucl. Chem., 1969, 31, 833 (luminescence).R.C. Holz and L.C. Thompson, Inorg. Chem., 1993, 32, 5251 (luminescence).D.F. Moser, L.C. Thompson, and V.G. Young, J. Alloys Compd., 2000, 303–304, 121.

Colour TV

Further Reading

J.R. McColl and F.C. Palilla, IAR, 177.G. Blasse, HRE, 4, 237.

Lasers

Further Reading

M.J. Weber, HRE, 4, 275.J. Hecht, The Laser Guidebook, 2nd edn., McGraw-Hill, New York, 1992.J. Fready, Kirk-Othmer Encyclopedia of Chemical Technology, 4th Edition,1995, vol. 15, 16.D.L. Andrews and A.A. Demydov (eds)., An Introduction to Laser Spectroscopy, Kluwer, Dordrecht,

2002.K. Kuriki, Y. Koike, and Y. Okamoto, Chem. Rev., 2002, 102, 2347 (plastic optical fibre lasers and

amplifiers).

ESR

Further Reading

E.M. Stephens, LPL, 181 (Gd ESR).A. Abragam and B. Bleaney, Electron Paramagnetic Resonance of Transition ions, Oxford, 1970.

NMR

Further Reading

F.A. Hart, CCC-1, 3, 1100 (NMR).J. Reuben and G.A. Elgarish, HRE, 4, 483 (NMR).J.H. Forsberg, HRE, 1996, 23, 1 (NMR of paramagnetic lanthanide compounds).C. Piguet and C.F.G.C. Geraldes, HRE, 2003, 33, 353 (solution structure of lanthanide-containing

coordination compounds by paramagnetic nuclear magnetic resonance).J. Reuben, Prog. Nucl. Magn. Reson. Spectrosc., 1975, 9, 1 (shift reagents).J. Reuben and G.A. Elgarish, HRE, 4, 483 (shift reagents).A.D. Sherry and C.F.G.C. Geraldes, LPL, 219 (shift reagents).M.R. Peterson Jr and G.H. Wahl Jr, J. Chem. Educ., 1972, 49, 790 (shift reagents).R.E. Sievers, Nuclear Magnetic Shift Reagents, Academic Press, 1972.F.A. Hart, CCC-1, 3, 1105 (shift reagents).M.F. Tweedle, LPL, 127 (MRI agents).R.B. Lauffer, Chem. Rev., 1987, 87, 901 (MRI agents).

SPH SPH


244 Bibliography

P. Caravan, J.J. Ellison, T.J. McMurry and R.B. Lauffer, Chem. Rev., 1999, 99, 2293 (MRI agents).A. Merbach and E. Toth (eds) The Chemistry of Contrast Agents in Medical Magnetic Resonance

Imaging, Wiley, Chichester, 2001.M.P. Lowe, Aust. J. Chem., 2002, 55, 551 (‘new generation’ MRI contrast agents).S. Aime, M. Botta and E. Terreno, Adv. Inorg. Chem. Radiochem., 2005, 57, 173 (MRI agents).L. Helm, G.M. Nicolle and A.E. Merbach, Adv. Inorg. Chem. Radiochem., 2005, 57, 327 (water and

proton exchange).

General References

T. Viswanathan and A. Toland, J. Chem. Educ., 1995, 72, 945 (chiral shift reagent).S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991, p. 64 (MRI agents).J.K.M. Sanders and D.H. Williams, Chem. Commun., 1970, 422 (shift reagents).

Lanthanide Biochemistry

Further Reading

J.-C.G. Bunzli, LPL, 219 (bioprobes).W de W. Horrocks and D.R. Sudnick, Acc. Chem. Res., 1981, 14, 384 (bioprobes).J. Reuben, HRE, 4, 515 (bioprobes).J.R. Ascenso and A.V. Xavier, SPL, 501 (bioprobes).F.S. Richardson, Chem. Rev., 1982, 82, 541 (bioprobes).C.H. Evans, Biochemistry of the Elements, Volume 8; Biochemistry of the Lanthanides, Plenum, New

York, 1990.J.R. Duffield, D.M. Taylor and D.R. Williams, HRE, 1991, 18, 591 (biochemistry).J. Burgess, Chem. Soc. Rev., 1996, 25, 85 (Sc, Y, lanthanides in living systems).J.M. Harrowfield, in A. Sigel and H. Sigel (eds), Metal Ions in Biological Systems, Marcel Dekker,

2003, vol. 40, 106 (interrelations of lanthanides with biosystems).G.R. Motson, J.S. Fleming and S. Brooker, Adv. Inorg. Chem., 2004, 55, 361 (Ln complexes as

luminescent biolabels).H. Tsukube, S. Shinoda and H. Tamiaki, Coord. Chem. Rev., 2002, 226, 227 (chiral recognition and

sensing of biological substrates).


Further Reading

F.G.N. Cloke, COMC-2, 3, 1 (zerovalent compounds).F.T. Edelmann,COMC-2, 3, 11 (all organometallics except zerovalent compounds).T.J. Marks and R.D. Ernst, COMC-1, 3, 173 (review).W.J. Evans, Adv. Organomet. Chem., 1985, 24, 131 (review).R.D. Kohn, G. Kociok-Kohn and H. Schumann, EIC-1, 3618 (review).H. Schumann and I.L. Fedushkin, EIC-2, 4878 (review).C.J. Schaverien, Adv. Organomet. Chem., 1994, 36, 283 (review).M.N. Bochkarev, L.N. Zakharov and G.S. Kalinina, Organoderivatives of Rare Earth Elements,

Kluwer, Dordrecht, 1995.W.A. Herrmann (ed.), Organolanthanoid Chemistry: Synthesis, Structure, Catalysis, in Topics in

Current Chemistry, Springer-Verlag, Berlin, 1996, vol. 179.

SPH SPH


Bibliography 245

H. Schumann, J.A. Meese-Marktscheffel and L. Essar, Chem. Rev., 1995, 95, 865 (review oforganometallic π -complexes).

G. Bombieri and G. Paolucci, HRE, 1998, 25, 265 (organometallic π -complexes).F.T. Edelmann, Angew. Chem., Int. Ed. Engl., 1995, 34, 2466 (cyclopentadienyl-free organolanthanide

chemistry).F.T. Edelmann, D.M.M. Freckmann and H. Schumann, Chem. Rev., 2002, 102, 1851 (non-

cyclopentadienyl organolanthanide complexes).S. Arndt and J. Okuda, Chem. Rev., 2002, 102, 1953 [mono(cyclopentadienyl) complexes].M.N. Bochkarev, Chem. Rev., 2002, 102, 2089 (arene–lanthanide compounds).F.G.N. Cloke, Chem. Soc. Rev., 1993, 22, 17 (zerovalent state).

General References

W.J. Evans and B.L. Davis, Chem. Rev., 2002, 102, 2119 ([(C5Me5)3Ln] systems).

7 The Misfits: Scandium, Yttrium and Promethium

These elements are covered, along with the lanthanides, in Section 39 of the Gmelin Handbook, mostof which extends to cover the 1980s.

Further Reading

R.C. Vickery, The Chemistry of Yttrium and Scandium, Pergamon, 1960; CIC, 3, 329.C.T. Horowitz (ed), Scandium, Academic Press, 1973 (comprehensive book to that point).S.A. Cotton, EIC-2, p 4838.F.A. Hart, CCC-1, 3, 1059 (coordination chemistry of Sc and Lanthanides).S.A. Cotton, CCC-2, 3, 93 (coordination chemistry of Sc and Lanthanides).S.A. Cotton, Polyhedron, 1999, 18, 1691 (Sc review).S.A. Cotton, Comments Inorg. Chem., 1999, 21, 165 [coordination number of Sc3+(aq)].F. Weigel, Chem. Zeit., 1978, 108, 339 (comprehensive review of Pm).G.E. Boyd, J. Chem. Educ., 1959, 36, 3 (Pm).E.J. Wheelwright (ed.), Promethium Technology, American Nuclear Society, 1973.S.A. Cotton, Educ. in Chem., 1999, 36, 96 (discovery of Pm, and its chemistry).

General References

P.S. Coan, L.G. Hubert-Pfalzgraf, and K.G. Caulton, Inorg. Chem., 1992, 31, 1262 [Y5O(OiPr)13

NMR].W.R. Wilmarth, R.G. Haire, Y.P. Young, D.W. Ramey, and J.R. Peterson, J. Less Common Metals (now

J. Alloys Compd)., 1988, 141, 275; W.R. Wilmarth, G.M. Bequn, R.G. Haire, and J.R. Peterson,J. Raman Spectrosc., 1988, 19, 271 (halides and oxide of Pm).

G.R. Willey, M.T. Lakin, N.W. Alcock, J. Chem. Soc., Chem. Commun., 1992, 1619 (Sc crown ethercomplex).

J.S. Ghotra, M.B. Hursthouse, and A.J. Welch, J. Chem. Soc., Chem. Commun., 1973, 669 (Scsilylamide).

J. Arnold, C.G. Hoffman, D.Y. Dawson, and F.J. Hollander. Organometallics, 1993, 12, 3646 (Sccomplexes).

SPH SPH


246 Bibliography


Further Reading

S. Kobayashi, Eur. J. Org. Chem., 1999, 15; S. Kobayashi, M. Sugiura, H. Kitagawa and W.W.-L.Lam, Chem. Rev., 2002, 102, 2227 (lanthanide triflates in organic synthesis).

S. Kobayashi and K. Manabe, Pure Appl. Chem., 2000, 72, 1373. (‘Green’ Lewis Acid catalysis inorganic synthesis).

H.B. Kagan, Tetrahedron, 2003, 59, 10351 (SmI2 overview).G.A. Molander and C.R. Harris, Chem. Rev., 1996, 96, 307; Tetrahedron, 1998, 54, 3321 (SmI2 in

sequencing reactions).A. Dahlen and G. Hilmersson, Eur. J. Inorg. Chem., 2004, 3393 (SmI2-mediated reductions – influence

of various additives).G.A. Molander and J.A.C. Romero, Chem. Rev., 2002, 102, 2161 (lanthanocene catalysts in selective

organic synthesis).J. Inanaga, H. Furuno and T. Hayano, Chem. Rev., 2002, 102, 2211 (asymmetric catalysis and ampli-

fication with chiral lanthanide complexes).K. Mikami, M. Terada and H. Matsuzawa, Angew. Chem., Int. Ed. Engl., 2002, 41, 3554 (asymmetric

catalysis by lanthanide complexes).H.C. Aspinall and N. Greeves, J. Organomet. Chem., 2002, 647, 151 (asymmetric synthesis).M. Shibasaki and N. Yoshikawa, Chem. Rev., 2002, 102, 2187 (lanthanide complexes in multifunc-

tional asymmetric catalysis).F. Helion and J.L. Namy, J. Org. Chem., 1999, 64, 2944; M.-I. Lannou, F. Helion and J.-L. Namy,

Tetrahedron, 2003, 59, 10551 (uses of mischmetal in organic synthesis).D. Klomp, T. Maschmeyer, U. Hanefeld and J.A. Peters, Chem. Eur. J., 2004, 10, 2088; S.-I. Fukuzawa,

N. Nakano and T. Saitoh, Eur. J. Org. Chem., 2004, 2863 (Meerwein–Ponndorf–Verley reaction).

General References

W.J. Evans, J.D. Feldman, and J.W. Ziller, J. Am. Chem. Soc., 1996, 118, 4581 [CeCl3(H2O)].T. Imamoto, Lanthanides in Organic Synthesis, Academic Press., 1994.S. Kobayashi (ed.), Lanthanides: Chemistry and Use in Organic Synthesis, Springer-Verlag,

Heidelberg, 1999.G.A. Molander, Chem. Rev., 1992, 92, 29 (application of lanthanide reagents in organic synthesis).


Further Reading

G.T. Seaborg and W.D. Loveland, The Elements beyond Uranium, Wiley, 1990.G.T. Seaborg, HRE, 1991, 18, 1 (origin of the actinide concept).M.S. Fred, ACE, 2, 1196 (oxidation states).J.J. Katz, L.R. Morss and G.T. Seaborg, ACE, 2, 1133, 1142 (oxidation states).L.J. Nugent, J. Inorg. Nucl. Chem., 1975, 37, 1767; MTP2, 7, 195 (oxidation states).L.R. Morss, HRE, 1991, 18, 239 (reduction potentials).T.W. Newton, APU, 24 (redox chemistry of Pu).Z. Kolaria, HAC, 3, 431 (extraction).Z. Yong-jun, HAC, 3, 469 (extraction).L.I. Katzin and D.C. Sonnenberger, ACE, 1, 46 (extraction of thorium).F. Weigel, ACE, 1, 196 (extraction of uranium).

SPH SPH


Bibliography 247

H.W. Kirby, ACE, 1, 112 (extraction of protactinium).F. Weigel, ACE, 1, 173 (separation of isotopes).S. Villani, Isotope Separation, American Nuclear Society, 1976; Uranium Enrichment, Springer-

Verlag, 1979.P. Pyykko, Adv. Quantum Chem., 1978, 11, 353 (relativistic effects).K.S. Pitzer, Acc. Chem. Res., 1979, 12, 271 (relativistic effects).P. Pyykko, Acc. Chem. Res., 1979, 12, 276 (relativistic effects).P. Pyykko, Chem. Rev., 1988, 88, 563 (relativistic effects).K. Balasubramanian, HRE, 1991, 18, 29 (relativistic effects).N. Kaltsoyannis. J. Chem. Soc., Dalton Trans., 1997, 1 (relativistic effects).B.A. Hess, Relativistic Effects in Heavy Element Chemistry and Physics, Wiley, 2002.N. Kaltsoyannis, Chem. Soc. Rev., 2003, 32, 9 (computational actinide chemistry, including relativistic

effects).

10 Binary Compounds

Further Reading

For uranium, consult GMEL, Supplement No. 55, Main Volume 1936, Supplements from 1975;A1–A7 (element), B2 (alloys), C1–C14 (compounds); D1–D4 (solution), E1–E2 (coordinationcompounds).

R.G. Haire and L. Eyring, HRE, 1991, 18, 413 (oxides).D. Brown, Halides of Lanthanides and Actinides, Wiley, 1968; MTP-1, 1, 87 (actinide halides).J.M. Haschke and R.G. Haire, APU, 212 (Pu binary compounds).

General References

S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991, p.95 (laser separation).D.D. Ensor, J.R. Peterson, R.G. Haire, and J.P. Young, J. Inorg. Nucl. Chem., 1981, 43, 2425 (EsF3).J.P. Young, R.G. Haire, J.R. Peterson, D.D. Ensor, and R.L. Fellows, Inorg. Chem., 1981, 20, 3979

(EsCl3).J.H. Burns, J.R. Peterson, and R.D. Baybarz, J. Inorg. Nucl. Chem., 1973, 35, 1171 (CfCl3).J.R. Peterson and J.H. Burns, J. Inorg. Nucl. Chem., 1973, 35, 1525 (CmCl3).Iu. V. Drobyshevskii, N.N. Prusakov, V.F. Serik, and V.B. Sokolov, Radiokhimiya, 1980, 22, 591

(claim for AmF6).J.-C. Berthet, M. Nierlich, and M. Ephritikhine, Chem. Commun., 2004, 870 (UO2I2).J.C. Taylor, Coord. Chem. Rev., 1976, 20, 197 (uranium halides and oxyhalides).A.F. Wells, Structural Inorganic Chemistry, Clarendon Press, Oxford, 3rd edn, 1962.

11 Co-ordination Chemistry of the Actinides

Further Reading

G.L. Soloveichik, EIC-1, 2 (complexes).D.W. Keogh, EIC-2, p. 2 (complexes).K.W. Bagnall, CCC-1, 3, 1129 (complexes).C.J. Burns, M.P. Neu, H. Boukhalfa, K.E. Gutowski, N.J. Bridges and R.D. Rogers, CCC-2, 3, 189

(complexes).

SPH SPH


248 Bibliography

Uranium

Further Reading

E.H.P. Cordfunke, The Chemistry of Uranium, Elsevier, 1969.R.G. Denning et al., LAS, 313; Mol. Phys., 1979, 37, 1109 (uranyl ion).W.R. Wadt, J. Am. Chem. Soc., 1981, 313, 6053 (uranyl ion).P. Pykko, L.J. Laakkonen, and K. Tatsumi, Inorg. Chem., 1989, 28, 1801 (uranyl ion).R.G. Denning, Struct. Bonding., 1992, 79, 215 (review on uranyl ion).U. Castellato, P.A. Vigato, and M. Vidali, Coord. Chem. Rev., 1981, 36, 183 (nitrate complexes).D. Brown, MTP-2, 7, 111 (nitrate complexes).D. Brown, MTP-1, 7, 87 (halide complexes).K.W. Bagnall, MTP-1, 7, 139 (thiocyanate complexes).R.K. Agarwal, H. Agarwal and K. Arora, Rev. Inorg. Chem., 2000, 20, 1 (thorium complexes of

neutral O-donors).K. Arora, R.C. Goyal, D.D. Agarwal and R.K. Agarwal, Rev. Inorg. Chem., 1999, 18, 283 (uranium

complexes of neutral O-donors).D.L. Clark, D.E. Hobart and M.P. Neu, Chem. Rev., 1995, 95, 25 (actinide carbonate complexes).M.P. Wilkerson, C.J. Burns, H.J. Dewey, J.M. Martin, D.E. Morris, R.T. Paine and B.L. Scott, Inorg.

Chem., 2000, 39, 5277 [uranium(VI) alkoxides and oxo alkoxides].W.G. Van Der Sluys and A.P. Sattelberger, Chem. Rev., 1990, 90, 1027 (alkoxides).W.G. Van Der Sluys, C.J. Burns, J.C. Huffman, and A.P. Sattelberger, J. Am. Chem. Soc., 1988, 110,

5924; J.M. Berg, D.L. Clark, J.C. Huffman, D.E. Morris, A.P. Sattelberger, W.E. Streib, W.G. VanDer Sluys and J.G. Watkin, J. Am. Chem. Soc., 1992, 114, 10811 (actinide aryloxides).

D.C. Bradley, R.C. Mehrotra, I.P. Rothwell, and A. Singh, Alkoxo and Aryloxo Derivatives of Metals,Academic Press, San Diego, 2001.

J-C. Berthet and M. Ephritikhine, Coord. Chem. Rev., 1998, 178–180, 83 (amides).

General References

J. Marcalo and A. Pires de Matos, Polyhedron, 1989, 8, 2431 (amides).

Other Elements

Further Reading

GMEL, Supplement No. 40, Main Volume 1940; Supplement 1981 (actinium).H.W. Kirby, ACE, 1, 14 (actinium).GMEL, Supplement No. 44, Main Volume 1955; Supplements A1–A4 (element), C1–C7 (com-

pounds), D1–D2 (spectra), E (coordination compounds) (thorium).L.I. Katzin and D.C. Sonnenberger, ACE, 1, 41 (thorium).I. Santos, A.P. de Matos, and A.G. Maddox, Adv. Inorg. Chem. Radiochem., 1989, 34, 65 (thorium).GMEL System No. 51, Main Volume 1942, 2 supplements 1977 (protactinium).D. Brown, AIP, 343; Adv. Inorg. Chem. Radiochem., 1969, 12, 1 (protactinium).J.A. Fahey, ACE, 1, 443 (neptunium).S.K. Patel, Coord. Chem. Rev., 1978, 25, 133 (neptunium).G.T. Seaborg, AIP, 1 (plutonium).F. Weigel, J.J. Katz and G.T. Seaborg, ACE, 1, 499 (plutonium).J.M. Cleveland, The Chemistry of Plutonium, American Nuclear Society, 1979.

SPH SPH


Bibliography 249

W.T. Carnall and G.R. Choppin (eds), ACS Symp. Series, 1983, vol. 216 (plutonium).C.J. Burns and A.P. Sattelberger, APU, 61 (organometallic and coordination chemistry of Pu).M.P. Neu, C.E. Ruggiero and A.J. Francis, APU, 170 (bioinorganic chemistry of Pu).S.B. Clark and C. Delegard, APU, 118 (chemistry of Pu in concentrated solutions).W.W. Schulz and R.A. Pennemann, ACE, 2, 887 (americium).J.D. Navratil, W.W. Schulz, and G.T. Seaborg, J. Chem. Educ., 1990, 67, 15 (americium).N.M. Edelstein, J.D. Navratil and W.W. Schulz, Americium and Curium Chemistry and Technology,

Reidel, 1985.E.K. Hulet and D.D. Bode, MTP-1, 7, 1 (nuclear waste processing).F. Weigel, J.J. Katz and G.T. Seaborg, ACE, 1, 505 (nuclear waste processing).K. Czerwinski, in APU, 77 (Pu separations).R.J. Silva and H. Nitsche, APU, 89 (Pu environmental chemistry).

General References

P.I. Artiukhin, V.I. Medcedovskii, and A.D. Gel’man, Radiokhimiya, 1959, 1, 131; Zh. Neorg. Khin.,1959, 4, 1324 (plutonium disproportionation) and self-reduction).

A. Von Grosse, J. Am. Chem. Soc., 1934, 56, 2501 (determination of atomic mass of protactinium).S.A. Cotton, Lanthanides and Actinides, Macmillan, 1991 (plutonium).

12 Electronic/Magnetic Properties of the Actinides

Further Reading

J.L. Ryan, MTP1, 7, 323 (UV–visible spectra).W.T. Carnall and H.M. Crosswhite, ACE, 2, 1235 (UV–visible spectra).S. Ahrland, CIC, 5, 473 (UV–visible spectra).J.P. Hessler and W.T. Carnall, LAS, 349 (UV–visible spectra).N.M. Edelstein and J. Goffart, ACE, 2, 1361 (magnetism).J.W. Gonsalves, P.J. Steenkamp, and J.G.H. Du Preez, Inorg. Chim. Acta., 1977, 21, 167; S. Afr.

J. Chem., 1977, 30, 62 (magnetism).B.N. Figgis and M.A. Hitchman, Ligand Field Theory and Its Applications, Wiley, 2000, 311–324

(magnetism and UV–visible spectra).

General References

M. Hirose et al., Inorg. Chim. Acta, 1988, 150, L93 (magnetism).J.L. Ryan and W.E. Keder, Adv. Chem. Ser., 1967, 71, 335 (UV–visible spectra).M.J. Sarsfield, H. Steele, M. Helliwell, and S.J. Teat, Dalton Trans., 2004, 3443 (UV–visible spectra).B.C. Lane and L.M. Venanzi, Inorg. Chim. Acta, 1969, 3, 239 (magnetism).J.G.H. DuPreez and B. Zeelie, Inorg. Chim. Acta, 1989, 161, 187 (UV–visible spectra).D.M. Gruen and R.L. Macbeth, J. Inorg. Nucl. Chem., 1959, 9, 297 UV–visible spectra).J.R. Peterson et al., Inorg. Chem., 1986, 25, 3779 (UV-visible spectra).


Further Reading

T.J. Marks, HAC, 4, 491; ACE, 2, 1588.

SPH SPH


250 Bibliography

T.J. Marks and A. Streitweiser, ACE, 2, 1547.T.J. Marks and I.L. Fragala (eds), Fundamental and Technological Aspects of Organo f-element

Chemistry, Reidel, 1985.T.J. Marks and R.D. Ernst, COMC-1, 3, 173.F.T. Edelmann, COMC-2, 1995, 4, 10.D.L. Clark and A.P. Sattelberger, EIC-1, 19.A.P. Sattelberger and D.L. Clark, EIC-2, p. 33.C.J. Burns and A.P. Sattelberger, APU, 61 (organometallic and coordination chemistry of Pu).

General References

T.J. Marks, A.M. Seyam, and J.R. Kolb, J. Am. Chem. Soc., 1973, 95, 5529 (UCp3X).T.J. Marks, and W.A. Wachter, J. Am. Chem. Soc., 1976, 98, 703 (ThCp3X).A. Streitweiser, Jr, and U. Mueller-Westerhoff, J. Am. Chem. Soc., 1968, 90, 7364 [U(η8-C8H8)2].


Further Reading

B. Fricke, Structure and Bonding, 1975, 21, 89 (predicted properties).G.T. Seaborg and O.L. Keller Jr, ACE, 2, 1629.A. Ghiorso, AIP, 23.R.J. Silva, ACE, 2, 1103.E.K. Hulet et al., J. Inorg. Nucl. Chem., 1980, 42, 79 (atom-at-a-time chemistry).K. Kumar, Superheavy Elements, Adam Hilger, Bristol and New York, 1989.G.T. Seaborg and W.D. Loveland, The Elements Beyond Uranium, Wiley-Interscience, 1990.S.A. Cotton, Chem. Soc. Rev., 1996, 25, 219.D.M. Hofman and D.M. Lee, J. Chem. Educ., 1999, 76, 332 (atom-at-a-time chemistry).S. Hofmann, On Beyond Uranium, Taylor and Francis, London, 2002.M. Schaedel (ed.), The Chemistry of Superheavy Elements, Kluwer, 2003.

General References

G.T. Seaborg, J. Chem. Educ., 1969, 46, 631 (regions of nuclear stability).G. Herrmann, Angew. Chem., Int. Ed. Engl., 1988, 27, 1421 (synthesis of element 107, now called

bohrium).J.V. Kratz, J. Alloys Compd., 1994, 213–214, 22 (apparatus for studying volatile halides).G.T. Seaborg, J. Chem. Soc., Dalton Trans., 1996, 3904 (Periodic Table).

Element 112

S. Hofmann, V. Ninov, F.P. Hessberger, P. Armbruster, H. Folger, G. Munzenberg, J. Schott, A.G.Popeko, A.V. Yeremin, S. Saro, R. Janik and M. Leino, Z. Phys. A. Hadrons Nucl., 1996, 354, 229.

Element 113

K. Morita et al, J. Phys. Soc. Jpn., 2004, 73, 2593.

SPH SPH


Bibliography 251

Element 114

Yu.Ts. Oganessian, A.V. Yeremin, A.G. Popeko, S.L. Bogomolov, G.K. Buklanov, N.L. Chelnokov,V.I. Chepigin, B.N. Gikal, V.A. Gorshkov, G.G. Gulbekian, M.G. Itkis, A.P. Kabachenko, A.Y.Lavrentev, O.N. Malyshev, J. Rohac, R.N. Sagadiak, S. Hofmann, S. Saro, G. Giardina andK. Morita, Nature, 1999, 400, 242.

Element 115

Yu.Ts. Oganessian, et al., Phys. Rev. C., 2004, 69, 021601.

Elements 116 and 118

V. Ninov, K.E. Gregorich, W. Loveland, A. Ghiorso, D.C. Hoffman, D.M. Lee, H. Nitsche, W.J.Swiatecki, U.W. Kirbach, C.A. Laue, J.L. Adams, J.B. Patin, D.A. Shaughnessy, D.A. Strellis andP.A. Wilk, Phys. Rev. Lett., 1999, 83, 1104; P.J. Karol, H. Nakahara, B.W. Petley and E. Vogt, PureAppl. Chem., 2001, 73, 959 (retraction).

SPH SPH


252

SPH SPH

JWBK057-IND JWBK057-Cotton December 13, 2005 5:39 Char Count=

Index

Note: Figures and Tables are indicated by italic page numbers

Absorption spectraactinides 202–7lanthanides 64, 68

Abundance, lanthanides 2–3Acetylacetonate complexes

lanthanide 39–40, 113scandium 110, 113

Actinide alkyls 210Actinide aqua ions 174Actinide aryls 210Actinide binary compounds 155–72Actinide carbonyls 222Actinide complexes

coordination numbers 174stability 174–5

Actinide cyclooctatetraene dianion compounds219–21

Actinide cyclopentadienyls 211–15Actinide dihalides 167, 168Actinide halide complexes 181–3Actinide halides 155–69

structure types 158–9syntheses 156–8

Actinide hexahalides 158, 161, 163–4, 166,167

Actinide ions, reduction potentials 150–2Actinide nitrate complexes 180–1Actinide organometallic compounds 209–24Actinide oxides 169–70Actinide oxyhalides 165–6, 166, 167,

170–1Actinide pentahalides 158–9, 161, 165, 166Actinide pentamethylcyclopentadienyl

compounds 215–18Actinide tetrahalides 158, 159, 161–3, 166,

167Actinide thiocyanates 183Actinide trihalides 157, 158, 158, 163, 166,

167, 168–9Actinides

characteristics 149–50electron configurations 150ionization energies 151

isotopes 147occurrence 145oxidation states 149, 156, 157, 160, 169,

188, 190–1relativistic effects 152synthesis of 145–6

Actiniumchemistry 186–7electron configuration 150ionization energies 151isotope(s) 147, 186metal 186

Actinium halides 165Actinium ion(s)

radius 186reduction potential(s) 151, 186

Adiabatic demagnetization 65Alcohols, oxidation of 123Alkali metal hydrides, with cerium(III) chloride

130Alkenes, oxidation of 123Alkoxides

lanthanide 48–9, 55, 57, 101as organic catalysts 135

scandium 111–12uranium 185, 186yttrium 115

Alkylamideslanthanide 47–8, 52scandium 111

Alkylsactinide 210lanthanide 89–90, 99

Alloys, lanthanide-based 25Americium

chemistry 195–6electron configuration 150ionization energies 151isotope(s) 147, 195

Americium complexes 196Americium halides 167Americium ion(s), reduction potential(s) 151,

195


SPH SPH


254 Index

Amidesreduction of 125–6uranium 184–5

Ammine complexeslanthanide 43scandium 111

Aqua ionsactinides 174lanthanides 18–19, 37scandium 108–9

Aqueous stability constants, lanthanide ion(s)36, 108

Arene complexesactinide 221–2lanthanide 101–2

Arenes, oxidation of 121–2Aromatic compounds, oxidation of 121–3Aryloxides

lanthanide 48–9, 55as organic catalysts 134–5

uranium 185Aryls

actinide 210lanthanide 90–1, 100

Atomic number 1, 115Atomic radii, lanthanides 14, 14, 15, 24Atomization enthalpies, lanthanides 17Atractyligenin, synthesis of 124–5

Barbier reactions 126Barium

atomic radius 14atomization enthalpy 17

Bastnasite (ore) 3, 3Berkelium

electron configuration 150ionization energies 151isotope(s) 147

Berkelium ion(s), reduction potential(s)151

Berkelium trichloride, absorption spectrum206–7

Binary compoundsactinides 155–72lanthanides 25–31

2,2’-Bipyridyl complexes, lanthanide 43–4Bis(cyclopentadienyl) compounds

actinide 211, 214lanthanide 92, 100

Bis(pentamethylcyclopentadienyl) alkylsactinide 215lanthanide 96, 97

Bis(pentamethylcyclopentadienyl)thoriummethyl cation 216–17

Bohrium (element 107)chemistry 235electron configuration 230

isotopes 231synthesis of 228

Borides, lanthanide 31Born–Haber cycles 15Bridging ligands, lanthanide cyclopentadienyls

92, 93

Calcium ions, aqueous stability constants 36Californium

electron configuration 150ionization energies 151isotope(s) 147synthesis of 146

Californium halides 168Californium ion(s), reduction potential(s) 151Carbides, lanthanide 31Carbonate complexes, actinide 178, 192, 193Carbonyl compounds

addition of halogenoalkanes to 126reduction of 124–6, 134–5

Carbonylsactinide 222lanthanide 102–3

Carboxylic acids, redction of 125Catalysts

actinide compounds as 216–17, 218lanthanide compounds as 127–8, 132,

134–5, 135–8Cerium

abundance 2, 3atomic radius 14, 15, 25atomization enthalpy 17electron configuration 9, 10, 12ionization energies 12, 13

Cerium(IV) alkoxides 101Cerium(IV) ammonium nitrate (CAN), use in

organic chemistry 121, 122, 123Cerium(IV) ammonium sulfate (CAS), use in

organic chemistry 121, 122Cerium bis(cyclooctatetradiendiyl) complex

101Cerium(III) chloride

with alkali metal hydrides 130uses in organic chemistry 128–30

Cerium(IV) compounds/complexes 56–7, 101as oxidants 121–3

Cerium halides 15–17, 16, 26, 27–8, 28, 29Cerium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpies 15magnetic moment 65radius 14, 15, 109redox potentials 20

Cerium oxides 29Chemical separation, lanthanides 4

SPH SPH


Index 255

Chinese rare-earth ores 3Chromium ion(s), radii 109Cobalt ion(s), radii 109Colours in aqueous solution, lanthanide ions

66Complexes

lanthanide 37–46stability 35–6

Cooper pair (electron pair) 31Coordination chemistry

actinides 173–99lanthanides 35–60scandium 108–14

Coordination numbers 50–1actinide compounds and complexes 158,

174first-order effects 51formulae and 54lanthanide compounds and complexes 51,

109, 113lanthanide contraction and 53–4scandium 112–14second-order effects 51

Copper(II) ions, aqueous stabilityconstants 36

Crown ether complexeslanthanide 41–2scandium 109, 110

Crystal field (CF) splittinglanthanide ions 66uranium ions 203

Curiumelectron configuration 150ionization energies 151isotope(s) 147

Curium(III) chloride 168Curium ion(s), reduction potential(s) 151Cyclooctatetraene dianion compounds

actinide 219–21lanthanide 98–9see also Uranocene

Cyclooctatetraenediyls, lanthanide 100Cyclopentadienyls

actinideoxidation state (III) 214–15oxidation state (IV) 211–14oxidation state (V) 211oxidation state (VI) 211

lanthanideoxidation state (II) 100oxidation state (III) 91–4

scandium 114see also Pentamethylcyclopentadienyl

compounds

d-block metals, compared with lanthanides 7Darmstadtium (element 110) 228, 230, 232

Diels–Alder reaction 127hetero-Diels–Alder reaction 128imino-Diels–Alder reaction 133–4lanthanide compounds as catalysts 127–8,

132Dihalides

actinide 167, 168lanthanide 17–18, 28–9β-Diketonate complexeslanthanide 39–41, 52, 77–8, 127–8

as NMR shift reagents 77–8, 127Lewis base adducts 41, 52scandium 110

Diphenylphenoxides, lanthanide 49–50Disproportionation

lanthanide halides 16, 17neptunium ions in acid solution 190–1

Dithiocarbamates, lanthanide 46Dithiophosphates, lanthanide 46Dubnium (element 105)

chemistry 234electron configuration 230isotopes 231synthesis of 227

Dysprosiumabundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 13

Dysprosium halides 16, 26, 28–9, 28Dysprosium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpy 15magnetic moment 65radius 14, 15, 109redox potentials 20

EDTA complexes, lanthanide 42–3, 80Einsteinium


Einsteinium(III) chloride 168–9Einsteinium ion(s), reduction potential(s) 151Electromagnetic process

transactinides separated by 230, 233uranium isotopes separated by 149,

163–4Electron configurations

actinides and ions 150lanthanides and ions 9, 10, 12transactinides 230

Electron Paramagnetic Resonance (EPR)spectroscopy, lanthanide ions 82–3

SPH SPH


256 Index

Electronic spectraactinides 202–7hypersensitive transitions 68–9lanthanide ions 64, 66–9

Element 112 228, 230, 232Element 113 229, 230, 232Element 114 228, 230, 232Element 115 229, 230, 232Element 116 228, 230, 232Element 118 228, 229, 230, 232Energy level diagrams

lanthanide ions 63, 65–6uranium(IV) ion 204uranyl ion 175, 176

Enthalpy of atomization, lanthanides 17Enthalpy changes, lanthanides 14–19Enthalpy of formation, lanthanide halides 16Enthalpy of hydration, lanthanide ions 15Erbium

abundance 2, 3aqueous stability constants for

Ln3 ions 36atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10hydration enthalpy for ion 15ionization energies 13redox potentials of ions 20

Erbium halides 16, 26Erbium ion(s)

colour in aqueous solution 66energy level diagram 63magnetic moment 65radius 14, 15, 109

Euro banknotes 77Europium

abundance 2, 3atomic radius 14, 14, 15, 23, 24atomization enthalpy 17boiling point of metal 23, 24, 25electron configuration 9, 10ionization energies 12, 13

Europium aqua ion, stability 19Europium complexes

electronic spectra 66luminescence 74, 75as NMR shift reagents 77, 78

Europium halides 16, 26, 28–9, 28Europium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpies 15luminescence 70–2, 73magnetic moment 65radius 14, 15, 109redox potential 20

Europium oxides 29–30Extraction, lanthanides 3–4

f orbitals 10, 11lanthanide properties affected by 10–11

Fermiumchemistry 196–7electron configuration 150ionization energies 151isotope(s) 147synthesis of 146

Fermium ion(s), reduction potential(s) 151Fluorescence, uranyl complexes 202Fluorox process 161Fractional crystallization, lanthanides 4Friedel–Crafts reactions 131

Gadolinite (ore) 1Gadolinium

abundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10, 12ionization energies 12, 13

Gadolinium complexes, medical applications45, 78, 80–2

Gadolinium halides 16, 26, 28, 29Gadolinium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63EPR spectroscopy 82, 84hydration enthalpy 15magnetic moment 65radius 14, 15, 109redox potentials 20

Gadolinium texaphyrin 45, 82, 83Gas centrifugation, uranium isotopes separated

by 148, 164Gaseous diffusion, uranium isotopes separated

by 148, 164Graham’s Law of gaseous diffusion 148,

164Grob oxidative fragmentation 123Ground state(s)

actinide ions 202, 203, 204lanthanide ions 62, 65

Group I metals, compared withlanthanides 7

Hafniumatomic radius 14atomization enthalpy 17

Halide complexesactinide 159–60, 181–3, 192lanthanide 44, 46, 56scandium 109

SPH SPH


Index 257

Halidesactinide 155–69lanthanide 14–18, 25–9scandium 108yttrium 114see also Dihalides; Tetrahalides; Trihalides

Halogen organic compounds, reduction of 124Hassium (element 108)


Hess’s Law 16, 17Hetero-Diels–Alder reactions 128Hexahalides, actinide 158, 161, 163–4, 166,

167Holmium

abundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 13

Holmium halides 16, 26, 28Holmium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpy 15magnetic moment 65radius 14, 15, 107, 109redox potentials 20

Hund’s rules 62Hydrated salts

lanthanide 37–8scandium 109

Hydration enthalpies, lanthanide ions 15Hydride cyclopentadienyl complexes,

lanthanide 98Hydride pentamethylcyclopentadienyl

complexesactinide 217–18lanthanide 98

Hydrides, lanthanide 31–2Hydrogenation catalysts 135–6

Imino-Diels–Alder reactions 133–4Insect traps 76International Union of Pure and Applied

Chemistry (IUPAC), nomenclature oftransactinides 229

Ion-exchange chromatography, lanthanides4–5, 6

Ionic radii, lanthanides 14, 14, 15, 50, 107,109

Ionization energiesactinides 151lanthanides 12, 13

Iron(III) ions, aqueous stability constants 36Iron ion(s), radii 109Isopropoxides, lanthanide 48–9

Ketones, α-heterosubstituted, reductionof 124

Landé g-factor 64Lanthanide alkoxides 48–9, 55, 57, 101

as organic catalysts 135Lanthanide alkylamides 47–8, 52Lanthanide alkyls 89–90, 99Lanthanide alloys 25Lanthanide ammine complexes 43Lanthanide aqua ions 37, 57

stability 18–19Lanthanide–arene complexes 101–2Lanthanide aryloxides 48–9, 55

as organic catalysts 134–5Lanthanide aryls 90–1, 100Lanthanide binary compounds 25–31Lanthanide bis(cyclopentadienyl) compounds

92Lanthanide bis(pentamethylcyclopentadienyl)

methyls 96, 97Lanthanide borides 31Lanthanide carbides 31Lanthanide carbonyls 102–3Lanthanide complexes 37–46

coordination numbers 51–4luminescence 74–5NMR applications 77–82stability 35–6

Lanthanide contraction 7–8, 11–12coordination numbers and 53–4examples 39

Lanthanide crown ether complexes 41–2Lanthanide cyclooctatetraene dianion

compounds 98–9Lanthanide cyclooctatetraenediyls 100Lanthanide cyclopentadienyls 91–4, 100Lanthanide dihalides 28–9

enthalpies of formation 16properties 28–9stability 17–18structures 28syntheses 28

Lanthanide β-diketonate complexes 39–41, 52as Diels–Alder catalysts 127–8as NMR shift reagents 77–8, 127

Lanthanide EDTA complexes 42–3Lanthanide halide complexes 44, 46, 56Lanthanide halides 25–9

enthalpies of formation 16stability 14–18see also Dihalides; Tetrahalides; Trihalides

Lanthanide hydrated salts 37–8

SPH SPH


258 Index

Lanthanide hydride cyclopentadienyl complexes98

Lanthanide hydrides 25, 31–2Lanthanide ions

aqueous stability constants 36, 108colours in aqueous solution 66coordination numbers 51, 109electron configurations 10, 12electronic spectra 65–9energy level diagrams 63, 65–6enthalpies of hydration 15EPR spectroscopy 82–3ground state 62luminescence 69–77magnetic properties 62–5as probes in biological systems 83–4radii 14, 14, 15, 50, 107, 109redox potentials 19, 20, 107

Lanthanide isopropoxides 48–9, 57Lanthanide–lanthanide bonds 103Lanthanide metallocenes 98–9Lanthanide mono(pentamethylcyclopentadienyl)

methyls 96, 97Lanthanide N-donor complexes 43–4Lanthanide nitrate complexes 41Lanthanide nitrides 31Lanthanide O-donor complexes 38–9,

56–7Lanthanide organometallic compounds 11,

89–105organic reactions catalysed by 136–8+2 oxidation state 99–100+3 oxidation state 89–99+4 oxidation state 101

Lanthanide oxides 29–31superconducting oxides 30–1

Lanthanide oxysulfides 32Lanthanide pentamethylcyclopentadienyls

94–5Lanthanide phosphors (TV displays) 75–6,

114Lanthanide S-donor complexes 46Lanthanide shift reagents (LSRs) 77Lanthanide silylamides 47, 52, 83Lanthanide sulfides 32Lanthanide tetrahalides 27–8

enthalpies of formation 16stability 14–17

Lanthanide thiolates 50Lanthanide triflates, uses in organic chemistry

131–4Lanthanide trihalides 25–7

enthalpies of formation 16properties 27structures 26–7syntheses 25–6

Lanthanides

abundance 2, 3atomic radii 14, 14, 15, 24atomization enthalpies 17characteristics 2compared with d-block and s-block metals 7coordination chemistry 35–60

(+2) oxidation state 55(+4) oxidation state 56–7

early history 1–2electron configurations 9, 10, 12elements/metals, properties 23–4enthalpy changes for formation of

compounds 14–19extraction 3–4, 24–5ionic radii 14, 15, 50, 107, 109ionization energies 12, 13metals

boiling points 23, 24, 25extraction/synthesis of 24–5in organic chemistry 135–6properties 23–4, 25uses 25

occurrence in various ores 3organometallic chemistry 11oxidation states 10, 14, 18, 29position in Periodic Table 1, 2,

6–7separation of 4–6

Lanthanumabundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17boiling point of metal 24, 25electron configuration 9, 10, 12ionization energies 13

Lanthanum aqua ion, stability 19Lanthanum boride 31Lanthanum compounds/complexes,

coordination numbers 113Lanthanum halides 15–16, 16, 26, 28, 29Lanthanum ion(s)

aqueous stability constants 36, 108colour in aqueous solution 66hydration enthalpy 15magnetic moment 65radius 14, 15, 109, 186redox potential(s) 20, 107, 186

Lanthanum–nickel alloys 25, 31, 135Lanthanum nickel hydrides 25, 31, 136Lasers

lanthanide ions in 76–7uranium isotopes separated using 149, 164,

165Lawrencium

electron configuration 150ionization energies 151isotope(s) 147, 225

SPH SPH


Index 259

Lawrencium ion(s), reduction potential(s)151

Lighting applications, lanthanide ions 76Lithium aluminium hydride, with cerium(III)

chloride 130Luminescence

Euro banknotes 77lanthanide ions/complexes 69–77

‘antenna’ effects 73, 74applications to sensory probes 74–5in lasers 76–7lighting applications 76quenching 73televison phosphors 75–6

Lutetiumabundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10, 12ionization energies 12, 13

Lutetium compounds/complexes, coordinationnumbers 113

Lutetium halides 16, 26Lutetium ion(s)

aqueous stability constants 36, 108colour in aqueous solution 66hydration enthalpy 15magnetic moment 65radius 14, 15, 109redox potential(s) 20, 107

Magnetic moments, lanthanide ions 64–5Magnetic properties

lanthanide ions 62–5uranium compounds 207–8

Magnetic resonance imaging (MRI)78–9

contrast agents 45, 79–81Manhattan project 4, 149, 163Mannich reactions 132–3Medical applications, lanthanide compounds

45, 78, 80–2Meerwein–Ponndorf–Verley reaction 135Meitnerium (element 109), electron

configuration 228, 230, 231Mendelevium


Mendelevium ion(s), reduction potential(s)151

Metal organic chemical vapour deposition(MOCVD) applications 57

Metallacycles, actinide 215, 216, 219Metallic properties

actinides 186, 187, 188lanthanides 23–4

Metallocenesactinide 219–21lanthanide 98–9

Mischmetal 25, 32in organic reactions 124, 135

Monazite (ore) 3, 3, 4, 145Mono(pentamethylcyclopentadienyl)

complexesactinide 216lanthanide 96, 97

N-donor complexeslanthanide 43–4scandium 112

Neodymiumabundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 13

Neodymium halides 16, 26, 28–9, 28Neodymium ion(s)


Neodymium lasers 76–7Neptunium

electron configuration 150ionization energies 151isotope(s) 147, 188

Neptunium complexes 188–9Neptunium halides 166Neptunium ion(s), reduction potential(s) 151,

188Neptunium oxyhalides 166, 171Nickel/lanthanide hydride batteries 25, 31Nitrate complexes

actinide 180–1, 193lanthanide 41scandium 110

Nitrides, lanthanide 31Nitriles, redction of 125–6NMR applications, lanthanide complexes

77–82Nobelium

chemistry 196–7electron configuration 150ionization energies 151isotope(s) 147

Nobelium ion(s), reduction potential(s) 151Nuclear waste

disposal of 194processing of 179–80

SPH SPH


260 Index

O-donor complexeslanthanide 38–9scandium 109–10

Occurrenceactinides 145lanthanides 2–3

Oresactinide-containing 145, 148lanthanide-containing 1, 3

Organic chemistry 121–43Organocerium compounds 128–9Organometallic compounds

actinide 209–24lanthanide 11, 89–105, 136–8

Oxidation, organic compounds 121–3Oxidation states

actinides 149, 156, 157, 160, 169, 188,190–1

lanthanides 10, 14, 18, 29Oxide superconductors 30–1, 114Oxides

actinide 169–70lanthanide 29–31scandium 108

Oxyhalides, actinide 165–6, 166, 167, 170–1Oxysulfides, lanthanide 32

Paramagnetic compounds 62, 207Pentahalides, actinide 158–9, 161, 165,

166Pentamethylcyclopentadienyl compounds

actinidecationic alkyls 216–17oxidation state (III) 218oxidation state (IV) 215–18

lanthanide 94–5scandium 114

Perchlorate complexes, lanthanide 44Periodic Table 1, 2, 6–7, 107, 115

future developments 2351,10-Phenanthroline complexes, lanthanide

43–4Phosphide complexes, lanthanide 50Phthalocyanine complexes 45, 178–9Pinacols 125Pitchblende (ore) 145, 148Plutonium

aqueous chemistry 189–90coordination chemistry 191–3electron configuration 150environmental considerations 193–4ionization energies 151isotope(s) 147oxidation states 190–1

Plutonium carbonate complexes 192, 193Plutonium halide complexes 192Plutonium halides 166–7

Plutonium ion(s)radii 109reduction potential(s) 151, 190

Plutonium nitrate complexes 193Plutonium oxyhalides 167, 171Plutonium triflates 193Porphyrins and related compounds

lanthanide 44–5scandium 112uranyl 178

Praseodymiumabundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 12, 13

Praseodymium halides 16, 26, 27–8, 28, 29Praseodymium ion(s)

aqueous stability constants 36colour in aqueous solution 66electronic absorption spectra 64, 66energy level diagram 63hydration enthalpy 15magnetic moment 65radius 14, 15, 109redox potentials 20

Praseodymium oxides 29Promethium 115–17

abundance 2, 3, 115atomic radius 14, 15, 25electron configuration 9, 10ionization energies 13isotopes 115–16

Promethium halides 16, 26, 116Promethium hydrated salts 116–17Promethium ion(s)


Promethium oxide 116Protactinium

chemistry 187–8electron configuration 150extraction of 148ionization energies 151isotope(s) 147, 187metal 187

Protactinium halide complexes 187Protactinium halides 158–9, 165Protactinium ion(s), reduction potential(s)

151Protactinium oxyhalides 165–6, 170–1Pyridine complexes, lanthanide 43

SPH SPH


Index 261

Radioactive lanthanide isotopes, separationfrom uranium 4

Redox/reduction potentialsactinide ions 150–2, 188, 195lanthanide ions 19, 20, 107

Reduction, organic compounds 124–7,134–5

Reformatsky reactions 127Relativistic effects 152, 234Roentgenium (element 111) 228, 230, 232Russell-Saunders coupling scheme 61, 201Rutherfordium (element 104)


s-block metals, compared with lanthanides 7Samarium

abundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17boiling point of metal 23, 24, 25electron configuration 9, 10ionization energies 13

Samarium halides 16, 26, 28–9, 28Samarium(II) iodide

as reducing agent 124–7synthesis of 124

Samarium ion(s)aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpies 15magnetic moment 65radius 14, 15, 109redox potentials 20term symbol for ground state 62

Samarium oxides 30Samarium(II) pentamethylcyclopentadienyl

100reactions 100, 101

Samarium(III) pentamethylcyclopentadienyl94, 95

Samarium sulfide 32Scandium 107–14

metal 108Scandium alkoxides 111–12Scandium alkylamides 111Scandium aqua ions 108–9, 113Scandium compounds/complexes, coordination

numbers 112–14Scandium crown ether complexes 109, 110Scandium cyclopentadienyls 114Scandium halide complexes 109Scandium halides 108, 113Scandium hydrated salts 109

Scandium ion(s)aqueous stability constants 36, 108coordination number(s) 109, 109, 113radius 107, 109redox potential(s) 107

Scandium O-donor complexes 109–10Scandium organometallic compounds 114Scandium oxide 108, 113Scandium silylamide 111Scandium triflate, uses in organic chemistry

131, 132Seaborgium (element 106)


Separation, lanthanides 4–6Siderophore complexes, plutonium in 194,

195SILEX system 164Silylamides

lanthanide 47, 52scandium 111

Sodium borohydride, with cerium(III) chloride130

Solvent extraction, lanthanides 5–6Stability constants, lanthanide ions in aqueous

solution 36, 108Sulfides, lanthanide 32Superconductors, lanthanide oxides 30–1, 114Supercritical carbon dioxide, use in separation

method 6

Television displays, phosphors in 75–6, 114Terbium

abundance 2, 3atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 12, 13

Terbium complexes, luminescence 74–5Terbium halides 16, 26, 27–8Terbium ion(s)

aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpy 15luminescence 70, 71magnetic moment 65radius 14, 15, 109redox potentials 20

Terbium oxides 29Tetra(cyclopentadienyl)actinide compounds

211–12Tetrahalides

actinide 158, 159, 161–3, 165, 166lanthanide 14–17, 27–8

SPH SPH


262 Index

Texaphyrins, lanthanide 45, 81–2, 83Thiocyanates

actinide 183lanthanide 43, 44, 52, 53

Thiolateslanthanide 46, 50uranium 184

Thoracene, half-sandwich compound 221Thorium

electron configuration 150extraction of 147ionization energies 150, 151isotope(s) 147reactions 156

Thorium(IV) bis(pentamethylcyclopentadienyl)methyl cation 216–17

Thorium halides 159–60Thorium ion(s)

aqueous stability constants 36, 175reduction potential(s) 150, 151

Thorium nitrate complexes 180, 181Thorium(IV) oxide 169Thorium oxyhalides 170Thortveitite (ore) 107Thulium

abundance 2, 3atomic and ionic radii 14atomic radius 14, 15, 24atomization enthalpy 17electron configuration 9, 10ionization energies 13

Thulium halides 16, 26, 28–9, 28Thulium ion(s)


Titanium ion(s), radii 109Transactinides 225–36

attempts to isolate 226chemistry 234–5electron configurations 230future developments 235, 236identification of elements 230, 233isotopes, half-lives and decay modes 231–2nomenclature 229prediction of chemistry 233–4stability considerations 225–6synthesis of 226–9

Transition metals, compared with lanthanides7

Tributyl phosphate, in nuclear waste processing179, 180

Tricolour lamps 76

Trihalidesactinide 156, 157, 158, 165, 166, 167,

168–9lanthanide 16, 25–7

Triphenylscandium 114Tris(cyclopentadienyl) compounds

actinide 214–15lanthanide 92, 93

Tris(cyclopentadienyl)uranium chloride,reactions 212, 213

Tris(pentamethylcyclopentadienyl) systemssamarium compound 94–5uranium compound 219

Uraninite (ore) 148Uranium

electron configuration 150extraction of 148

other elements obtained from 107, 148fission products, lanthanides separated from

4ionization energies 151isotope(s) 147

separation of 148–9oxidation states 157, 160reactions 156

Uranium alkoxides 185, 186Uranium amides 184–5Uranium(III) bis(pentamethylcyclopentadienyl)

chloride 218Uranium compounds, magnetic properties

207–8Uranium halide complexes 181–3

temperature-dependent magneticsusceptibility 208

Uranium halides 160–3Uranium hexafluoride

isotope separation using 163–4, 165synthesis of 161

Uranium hydride 170Uranium imides 211, 212Uranium ion(s)

aqueous stability constants 36reduction potential(s) 151, 174

Uranium nitrate complexes 179, 180–1Uranium oxides 169–70Uranium oxyhalides 171Uranium thiolates 184Uranocene 219–20

bonding in 221Uranyl acetate 178Uranyl alkoxides 185Uranyl carbonate complexes 178Uranyl complexes 175–7

coordination numbers 177fluorescence 202structures 177, 178

SPH SPH


Index 263

Uranyl ion (UO22+)

absorption spectra 202, 203aqueous stability constants 36, 175bonding in 175, 176energy level diagram 175, 176

Uranyl nitrate 179in nuclear waste processing 180solubility in various solvents 179

Uranyl ‘superphthalocyanine’ complex178–9

Vanadium ion(s), radii 109

Xenotime (ore) 3, 3

Ytterbiumabundance 2, 3atomic radius 14, 14, 15, 23, 24atomization enthalpy 17boiling point of metal 23, 24, 25electron configuration 9, 10ionization energies 12, 13

Ytterbium halides 16, 26, 28–9, 28

Ytterbium ion(s)aqueous stability constants 36colour in aqueous solution 66energy level diagram 63hydration enthalpies 15magnetic moment 65radius 14, 15, 109redox potentials 20

Yttria 1Yttrium 114–15

abundance 2, 3electron configuration 10ionization energies 13

Yttrium alkoxides 115Yttrium aluminium garnet (YAG) 76Yttrium compounds 114–15Yttrium halides 114Yttrium ion(s)

aqueous stability constants 36, 108hydration enthalpy 15radius 107, 109redox potential(s) 20, 107

Yttrium oxide superconductors 30–1, 114

With kind thanks to Paul Nash for creation of the index.

lanthanide and actinide chemistry · ffx ffx jwbk057-fm jwbk057-cotton december 9, 2005 14:5 char...

Documents